PharmacoEconomics

The costs of diagnosing and treating sexually transmitted in low- and middle- income countries from 2006 to 2014: An updated systematic review --Manuscript Draft--

Manuscript Number: Full Title: The costs of diagnosing and treating sexually transmitted infections in low- and middle- income countries from 2006 to 2014: An updated systematic review Article Type: Systematic Review

Funding Information: United States Agency for International Not applicable Development (AID-674-A-12-00029)

Abstract: Background: Sexually transmitted infections (STIs) are co-factors for HIV and can cause significant morbidity. Expanding on a prior systematic review, we aimed to summarize recent literature on the costs of diagnosing and treating curable STIs in low- and middle-income countries (LMICs).

Methods: We conducted a systematic review using pre-established search strategies. Citations were eligible if published between 1 January 2006 and 31 December 2014 and if they contained provider-perspective cost information reflective of STI-related service provision in LMICs. We extracted all cost values and used regression analysis to explore determinants. Cost drivers were analyzed thematically.

Results: We identified 44 articles for inclusion; 24 (54.6%) represented Sub-Saharan Africa. We extracted 202 cost values; 72 (35.6%) characterized syndromic management approaches, 57 (28.2%) mobile outreach services. was a common focus (70 (34.7%)). Sixty-five (32.2%) cost values represented cost- effectiveness measures as compared to simple unit costs. The median for all cost values was (USD 2015) $10.90 (cost-effectiveness measures $115.88; unit costs $4.15). Regression analysis indicated that cost effective measures were lower in Africa than other continents. Comparing unit costs only, costs were lower for mobile services than fixed-site services. Thirty-seven (84.1%) articles discussed cost drivers. Service delivery mode was most common, followed by volume/scale, STI incidence/prevalence, drug costs and laboratory costs.

Conclusions: Many of the articles reviewed presented partial information. Efforts for accelerating action on STIs in LMICs would benefit from increased availability of information on STI prevalence, effective interventions to address STIs (diagnostic approaches and medications) and costing methodologies and outcomes. Corresponding Author: Naomi Lince-Deroche, PhD Health Economics and Epidemiology Research Office Johannesburg, SOUTH AFRICA Corresponding Author Secondary Information: Corresponding Author's Institution: Health Economics and Epidemiology Research Office Corresponding Author's Secondary Institution: First Author: Naomi Lince-Deroche, PhD First Author Secondary Information: Order of Authors: Naomi Lince-Deroche, PhD Rahma Leuner Calvin Chiu Andrea Teagle

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation Sharon Kgowedi Cynthia Firnhaber Order of Authors Secondary Information: Author Comments: To the Editors:

On behalf of my co-authors I am pleased to submit the attached manuscript titled: "The costs of diagnosing and treating sexually transmitted infections in low- and middle- income countries from 2006 to 2014: An updated systematic review," for consideration by the Editorial Board at PharmacoEconomics.

NLD and CC conceptualized and designed the study. NLD, CC, AT, RL, and SK implemented the study. NLD, RL, and CC conducted the analysis. NLD, CC, RL, and SK drafted the manuscript. All authors read and approved the final manuscript.

This work did not require ethical clearance as no human subjects were involved. We declare that we have no conflicting interests related to the research or its potential application.

This manuscript was previously submitted to Sexually Transmitted Diseases and reviewed by three peer reviewers. It was rejected. We feel that the comments/objections reflected a limited knowledge of health economics and the value of presenting cost data in manuscripts. Thus, we have not revised the manuscript based on the comments received.

We look forward to future communication with the journal.

Regards, Naomi Lince-Deroche Suggested Reviewers: David Lewis [email protected] International STI expert, knowledgeable of LMIC settings

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation Cover Letter

Health Economics and Epidemiology Research Office

Unit 2, 39 Empire Road Parktown, Johannesburg, 2193, South Africa Tel +27 (0)10 0017930

A division of the Wits Health Consortium (Pty) Ltd, a wholly owned subsidiary of the University of the Witwatersrand

12 February 2018

The Editors Sexually Transmitted Diseases

RE: Submission of manuscript

To the Editors:

On behalf of my co-authors I am pleased to submit the attached manuscript titled: “The costs of diagnosing and treating sexually transmitted infections in low- and middle-income countries from 2006 to 2014: An updated systematic review,” for consideration by the Editorial Board at PharmacoEconomics.

NLD and CC conceptualized and designed the study. NLD, CC, AT, RL, and SK implemented the study. NLD, RL, and CC conducted the analysis. NLD, CC, RL, and SK drafted the manuscript. All authors read and approved the final manuscript.

This work did not require ethical clearance as no human subjects were involved. We declare that we have no conflicting interests related to the research or its potential application.

This manuscript was previously submitted to Sexually Transmitted Diseases and reviewed by three peer reviewers. It was rejected. We feel that the comments/objections reflected a limited knowledge of health economics and the value of presenting cost data in manuscripts. Thus, we have not revised the manuscript based on the comments received.

We look forward to future communication with the journal.

Regards,

Naomi Lince-Deroche [email protected]

Manuscript Click here to download Manuscript Lince-Deroche. STI Dx and Tx Cost 2018.02.12.docx Click here to view linked References

1 1 Title: The costs of diagnosing and treating sexually transmitted infections in low- and 2 3 4 2 middle-income countries from 2006 to 2014: An updated systematic review 5 6 3 7 8 4 Authors: Naomi Lince-Deroche1, Rahma Leuner1, Calvin Chiu1, Andrea Teagle1, Sharon 9 10 1 2, 3 11 5 Kgowedi , Cynthia Firnhaber 12 13 6 14 15 16 7 Author affiliations: 17 18 8 1 Health Economics and Epidemiology Research Office, Department of Internal 19 20 21 9 Medicine, School of Clinical Medicine, Faculty of Health Sciences, University of 22 23 10 the Witwatersrand, Johannesburg, South Africa 24 25 11 2 Right to Care, Johannesburg, South Africa 26 27 28 12 3 Clinical HIV Research Unit, Department of Internal Medicine, Faculty of Health 29 30 13 Sciences, University of Witwatersrand, Johannesburg, South Africa 31 32 33 14 34 35 15 Corresponding author: Naomi Lince-Deroche, Health Economics and Epidemiology 36 37 38 16 Research Office, 39 Empire Rd, Parktown, Johannesburg, South Africa 2194, phone: 39 40 17 +27 10 010 0638, email: [email protected] 41 42 43 18 44 45 19 Word counts: Summary 30, Abstract 250, Text 2,999 46 47 20 Number of figures: 4 / tables: 2 48 49 50 21 51 52 22 Running title: Systematic review of STI costs in LMICs 53 54 55 56 57 58 59 60 61 62 63 64 65 23 1 2 24 Acknowledgments 3 4 5 25 The authors gratefully acknowledge contributions from Dr Fern Terris-Prestholt 6 7 26 during the preparation of this manuscript. 8 9 10 27 This study was made possible by the generous support of the American people 11 12 28 through the United States Agency for International Development (USAID), award number 13 14 AID-674-A-12-00029. The contents are the responsibility of the Health Economics and 15 29 16 17 30 Epidemiology Research Office, a Division of the Wits Health Consortium (Pty) Ltd and do 18 19 31 not necessarily reflect the views of USAID or the United States Government. 20 21 22 32 23 24 33 Key points: 25 26 27 34  A systematic review of literature published from 2006-2014 of sexually 28 29 35 transmitted infection (STI) screening and treatment costs in low- and middle- 30 31 32 36 income countries found that data are lacking on specific interventions and 33 34 37 costing methodologies. 35 36 37 38  A meta-analysis of available data produced a median cost value of (USD 38 39 39 2015) $10.90 (cost-effectiveness measures $115.88; unit costs $4.15). 40 41 42 40  Regression analysis indicated that cost effective measures were lower in 43 44 41 Africa than other continents. 45 46 42  Considering cost drivers, service delivery mode was most commonly 47 48 49 43 mentioned, followed by volume/scale, STI incidence/prevalence, drug costs 50 51 44 and laboratory costs. 52 53 54 55 56 57 58 59 60 61 62 63 64 1 65 45 Abstract 1 2 46 Background: Sexually transmitted infections (STIs) are co-factors for HIV infection and can 3 4 5 47 cause significant morbidity. Expanding on a prior systematic review, we aimed to summarize 6 7 48 recent literature on the costs of diagnosing and treating curable STIs in low- and middle- 8 9 10 49 income countries (LMICs). 11 12 50 13 14 Methods: We conducted a systematic review using pre-established search strategies. Citations 15 51 16 17 52 were eligible if published between 1 January 2006 and 31 December 2014 and if they 18 19 53 contained provider-perspective cost information reflective of STI-related service provision in 20 21 22 54 LMICs. We extracted all cost values and used regression analysis to explore determinants. 23 24 55 Cost drivers were analyzed thematically. 25 26 27 56 28 29 57 Results: We identified 44 articles for inclusion; 24 (54.6%) represented Sub-Saharan Africa. 30 31 58 We extracted 202 cost values; 72 (35.6%) characterized syndromic management approaches, 32 33 34 59 57 (28.2%) mobile outreach services. Syphilis was a common focus (70 (34.7%)). Sixty-five 35 36 60 (32.2%) cost values represented cost-effectiveness measures as compared to simple unit 37 38 61 costs. The median for all cost values was (USD 2015) $10.90 (cost-effectiveness measures 39 40 41 62 $115.88; unit costs $4.15). Regression analysis indicated that cost effective measures were 42 43 63 lower in Africa than other continents. Comparing unit costs only, costs were lower for mobile 44 45 46 64 services than fixed-site services. Thirty-seven (84.1%) articles discussed cost drivers. Service 47 48 65 delivery mode was most common, followed by volume/scale, STI incidence/prevalence, drug 49 50 51 66 costs and laboratory costs. 52 53 67 54 55 68 Conclusions: Many of the articles reviewed presented partial information. Efforts for 56 57 58 69 accelerating action on STIs in LMICs would benefit from increased availability of 59 60 61 62 63 64 2 65 70 information on STI prevalence, effective interventions to address STIs (diagnostic 1 2 71 approaches and medications) and costing methodologies and outcomes. 3 4 5 72 6 7 73 Key Words: Economic, chlamydia, gonorrhea, syphilis, testing, treatment 8 9 10 74 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 3 65 75 1. Introduction 1 2 76 According to the World Health Organization (WHO) more than 30 types of bacteria, 3 4 5 77 viruses, and parasites are known to be transmitted through sexual conduct, and eight of these 6 7 78 are responsible for the greatest burden of sexually transmitted disease [1]. Globally each year 8 9 10 79 there are 357 million new infections with four curable sexually transmitted infections (STIs): 11 12 80 gonorrhea (Neisseria gonorrhoeae), chlamydia (Chlamydia trachomatis), syphilis 13 14 (Treponema pallidum) and trichomoniasis (Trichomonas vaginalis) [1]. The incidence of 15 81 16 17 82 curable STIs is highest in the Americas and Africa [2]. Four viral, non-curable infections— 18 19 83 HIV, hepatitis B, human papilloma virus (HPV) and herpes simplex virus (HSV) — also 20 21 22 84 contribute significantly to the global burden of STI-related disease [2]. 23 24 85 Infection with curable and non-curable STIs can cause significant discomfort to 25 26 27 86 individuals who experience symptoms, and all infected individuals – symptomatic and 28 29 87 asymptomatic alike – are at risk of more serious sequelae. Infection with certain STIs such as 30 31 32 88 syphilis and herpes can increase risk of HIV acquisition, and infection with chlamydia and 33 34 89 gonorrhea are significant causes of pelvic inflammatory disease (PID) and infertility in 35 36 90 women.[1] STI infection in pregnant women can also result in pregnancy complications and 37 38 39 91 neonatal infection which can result in serious morbidity and mortality [1,3]. 40 41 92 Control of STIs is a global priority addressed in the Sustainable Development Goals 42 43 44 93 [4]. Nonetheless STI management remains a major challenge, especially in low- and middle- 45 46 94 income countries (LMICs) where health system infrastructure is least developed [5]. There 47 48 49 95 are several strategies for STI diagnosis and treatment globally. The WHO has historically 50 51 96 recommended syndromic management (where treatment is provided for signs and symptoms 52 53 of STIs rather than known infection with a specific pathogen) for low-resource settings [6]. 54 97 55 56 98 However, in 2016, the World Health Organization (WHO) released new guidelines for 57 58 99 accelerating efforts to prevent and control STIs for the period 2016-2021, and 59 60 61 62 63 64 4 65 100 recommendations have shifted towards etiological management [7]. Laboratory-based testing 1 2 101 is available in better-resourced settings, and point-of-care and rapid testing solutions are 3 4 5 102 becoming increasingly more available for specific STIs including chlamydia, gonorrhea, and 6 7 103 human papillomavirus [8–10]. 8 9 10 104 In 2006, Terris-Prestholt et al published a systematic review summarizing the costs 11 12 13 105 and cost drivers of treating curable STIs in LMICs [11]. The review covered the period of 14 15 106 1980-2005 and included an analysis of factors contributing to costs – including those 16 17 18 107 associated with the process of providing services and the outcomes chosen to measure 19 20 108 impact. Given renewed commitments to achieving universal access to health care, including 21 22 23 109 the prevention and control of STIs, for this analysis we aimed to update and expand on the 24 25 110 prior systematic review by Terris-Prestholt et al. We summarize recent literature on the costs 26 27 111 of diagnosing and treating curable STIs in LMICs and discuss reported cost drivers. 28 29 30 31 112 32 33 113 2. Methodology 34 35 114 2.1 Systematic review 36 37 38 115 We conducted a systematic review using a detailed protocol. The initial citation 39 40 116 search was conducted in November-December 2013; it was updated in August 2015 and 41 42 43 117 April 2017. We searched PubMed, PopLine, J-STOR, and Google Scholar using pre- 44 45 118 established search strategies containing both free text and MeSH terms. We also emailed 46 47 48 119 experts in the field requesting possible articles. 49 50 120 Search terms included “sexually transmitted” and appropriate variations of each of the 51 52 121 following pathogens or diseases: chlamydia, gonorrhea, trichomoniasis, syphilis, bacterial 53 54 55 122 vaginosis, candida, chancroid, donovanosis, epididymitis, epididymo-orchitis, genital scabies, 56 57 123 genital ulcers, genital warts, herpes, lower abdominal pain, lymphogranuloma venereum, 58 59 60 124 pelvic inflammatory disease, pubic lice, reproductive tract infection, urethritis, urethral 61 62 63 64 5 65 125 discharge, and vaginal discharge. These disease terms were combined with cost terms and 1 2 126 terms referencing LMICs (as defined by The World Bank in 2013 [12]) via Boolean operators 3 4 5 127 to facilitate database searches. The full search strategy is available in Appendix A. 6 7 128 The search was limited to all publications published from 1 January 2006 to 31 8 9 10 129 December 2014. We allowed original research articles, editorials/commentary, reports or fact 11 12 130 sheets with new or current data, and conference proceedings, presentations, and posters. We 13 14 accepted material published in English, German, French, Portuguese, Spanish, Italian, or 15 131 16 17 132 Afrikaans. However, the searches were conducted in English. 18 19 133 Citations were eligible for selection if they provided cost information reflective of 20 21 22 134 service provision in LMICs. We excluded citations that did not include costs in the abstract or 23 24 135 full text. We also excluded citations that provided only patient costs. Eligible articles may 25 26 27 136 have included these costs, but had to also include provider perspective costs. 28 29 137 After retrieving the initial search results using all search engines, we removed 30 31 32 138 duplicates. Then two reviewers separately screened all titles and abstracts. Discrepancies in 33 34 139 decision making regarding inclusion at this stage were arbitrated by a third reviewer. We then 35 36 140 searched for the full text of articles deemed eligible based on the title or abstract. Ultimately, 37 38 39 141 if the full text could not be located, the citation was considered ineligible. 40 41 142 We extracted cost data from eligible, full-text sources using an Excel-based tool 42 43 44 143 designed for this analysis. All costs presented by the authors were included, including both 45 46 144 unit costs and cost effectiveness measures. This also included costs that may not have been 47 48 49 145 empirically derived by the authors, but which were presented in the publication. All cost 50 51 146 values were converted from their reported dollar or euro values into local currencies using 52 53 average exchange rates for the cost year [13] and then inflated to 2015 local prices using local 54 147 55 56 148 Consumer Price Indices [14]. Costs were then converted into US dollars using average 57 58 149 exchange rates for 2015 for ease of comparison [13]. If the country of origin for the cost 59 60 61 62 63 64 6 65 150 value was uncertain, we inflated the value to 2015 prices using the Consumer Price Index of 1 2 151 the United States under the assumption that this was a conservative lower-bound estimate. 3 4 5 152 For cost values where the cost year was not provided, we assumed the cost year to be the year 6 7 153 of publication. Finally, for cost values where only a range was given, we imputed the base 8 9 10 154 case value using the mean of the upper and lower bound estimate. 11 12 155 13 14 15 156 2.2 Regression analysis 16 17 157 Adapting the methods used in the original systematic review by Terris-Prestholt et al. 18 19 158 [11], we used Stata (v14, StataCorps LP, College Station, TX) to conduct regression analysis 20 21 22 159 to explore determinants of variation in the extracted cost values. The analysis considered: 23 24 160 intervention type (syndromic management versus etiological, mobile outreach versus facility- 25 26 27 161 based), where the intervention took place (Sub-Saharan Africa or other non-Sub-Saharan 28 29 162 LMICs), how the costs were estimated and presented (financial or economic, full or 30 31 32 163 incremental), and other study- and country-specific factors (number of intervention sites, year 33 34 164 of publication, and historical per capita gross domestic product (GDP) (PCGDP)). Historical 35 36 165 PCGDP was included to account for differences in country income levels which may have 37 38 39 166 affected the cost values. The number of individuals treated was not included in the model due 40 41 167 to high levels of missing data and collinearity with the number of sites investigated. 42 43 44 168 Finally, we conducted the regression analysis three times to explore the significance 45 46 169 of cost effectiveness outcomes versus unit costs. The first run of the analysis included the full 47 48 49 170 sample – with both cost effectiveness outcomes and unit costs. The second and third runs of 50 51 171 the analysis included either cost effectiveness outcomes or unit costs only. PCGDP was 52 53 excluded from the regression containing cost-effectiveness measures only due to collinearity 54 172 55 56 173 with the year of publication. 57 58 59 60 61 62 63 64 7 65 174 As in Terris-Prestholt’s review, due to large and influential cost outliers, the 1 2 175 dependent variable in all three analyses was defined as the natural log of the cost values. We 3 4 5 176 used a linear representation of all but one of the independent variables (number of 6 7 177 intervention sites), as in the review by Terris-Prestholt, again due to large and potentially 8 9 10 178 influential outliers. Finally, we opted to specify use of robust standard errors to correct for 11 12 179 any bias introduced by noted heteroscedasticity in the model. 13 14 180 15 16 17 181 3. Results 18 19 182 We identified 1,406 citations through the initial search, and eliminated 1,121 citations 20 21 22 183 in the first screening (see Figure 1). An additional 240 citations/articles were removed during 23 24 184 full text screening, most often because they did not contain cost values. Ultimately, 44 full 25 26 27 185 text articles were selected for inclusion in the review. 28 29 186 30 31 32 187 3.1 Summary of articles 33 34 188 Table 1 provides detailed information on the 44 papers selected for inclusion in the 35 36 189 review. Twenty-five of the 44 papers (56.8%) presented costs for services provided in LMICs 37 38 39 190 in Sub-Saharan Africa [15–39]. Three of those represented Sub-Saharan African countries 40 41 191 generally [35–37], and one included a Sub-Saharan African and a Latin American country 42 43 44 192 [38]. Six additional papers (11.4%) represented Latin America and the Caribbean exclusively 45 46 193 [40–45]. Ten papers (22.7%) presented costs from Asian countries [46–55]. Two were 47 48 49 194 European countries (4.5%) [56,57], and one modelling paper did not specify a country [58]. 50 51 195 The number of specific intervention locations represented in the evaluations ranged 52 53 196 from 0 to 61 (with the 0 representing a hypothetical, modelling exercise); 34.1% (n=15) of 54 55 56 197 the papers presented data representing one or two locations. The population sample sizes used 57 58 198 to determine the costs of diagnosis or treatment ranged widely – from 50 to 500,000 – with 59 60 61 62 63 64 8 65 199 modelling papers tending to have the larger sample sizes. The target populations represented 1 2 200 in the studies included antenatal clinic attendees (n=15, 34.1%), symptomatic individuals 3 4 5 201 presenting for care at an STI clinic or other location (n=11, 25.0 %), female sex workers 6 7 202 (n=5, 11.4%), the general population (n=3, 6.8%), and a range of other populations (n=10, 8 9 10 203 22.7%). Just two papers (4.6%) contained target populations of patients co-infected with 11 12 204 HIV. 13 14 Considering the cost data specifically, from the 44 papers, the study team extracted 15 205 16 17 206 202 cost values representing diagnosis and/or treatment of STIs. Most of the cost values 18 19 207 (n=124, 61.4%) represented services provided in Sub-Saharan African countries. Eighty 20 21 22 208 percent (n=161) represented costs from the health provider’s perspective; 19.3% (n=39) 23 24 209 represented societal costs (i.e. provider plus patient costs). Seventy-two (35.6%) of the cost 25 26 27 210 values represented syndromic management of STIs (versus etiological management), and 57 28 29 211 (28.2%) represented data from mobile outreach services (versus fixed-site services). Costs for 30 31 32 212 diagnosis and /or treatment of “general STIs” or “any STIs” were provided often (n=56, 33 34 213 27.7%); though costs for the individual diagnosis and treatment of syphilis were more 35 36 214 common (n=71 (35.1%)). 37 38 39 215 Sixty-five (32.2%) of the cost values represented cost-effectiveness measures, such as 40 41 216 the cost per HIV infection averted or the cost per disability-adjusted life year saved (DALY). 42 43 44 217 Figure 2 presents a break-down of the specific categorization of unit costs and cost- 45 46 218 effectiveness measures used. Unit costs made up the majority of the cost values (n=137, 47 48 49 219 67.8%). The two most common cost types were cost per case/person treated (n=66, 32.7%) 50 51 220 and cost per drug/test/supply (n=31, 15.3%). Eight (4.0%) of the cost values represented 52 53 treatment for newborns or mother-newborn pairs or the cost effectiveness of averting 54 221 55 56 222 newborn infection or low birth weight (data not shown). 57 58 59 60 61 62 63 64 9 65 223 Half (103 (51.0%)) of the cost values were derived using a “bottom-up” approach to 1 2 224 estimate costs meaning that the costs were calculated using various inputs derived from 3 4 5 225 observing actual service provision. In contrast, 96 (47.5%) of the cost values were derived 6 7 226 using an ingredients approach; these costs were calculated based on expected service delivery 8 9 10 227 norms or policy documents. Just three of the cost values (1.5%) represented “top-down” 11 12 228 approaches, meaning that total projects costs were divided by the outcome of interest to 13 14 determine the cost per outcome. 15 229 16 17 230 Seventy-eight (38.6%) of the cost values were derived from mathematical modelling 18 19 231 exercises (as compared to empirically collected data). Twenty-seven cost values (13.5%) 20 21 22 232 were full costs in that they included overhead in addition to direct costs such as staff time, 23 24 233 equipment, consumables and medications. Finally, 34 (16.8%) of the cost values represented 25 26 27 234 economic costs, which may have included the market value of goods and services (if different 28 29 235 from expenditure), the value of donated goods and services, etc., as compared to financial 30 31 32 236 costs, which represented expenditure only. 33 34 237 35 36 238 3.2 Cost outcomes 37 38 39 239 The median of the 202 cost values was USD $10.90 (Interquartile Range (IQR) $2.21- 40 41 240 $70.47). Due to wide variation in the methodologies used to estimate the costs as well as 42 43 44 241 differences in the outcome measures, the overall median cost should be interpreted with 45 46 242 caution. To illustrate, for cost effectiveness measures only, the median cost was USD 47 48 49 243 $115.88 (IQR $18.43-$404.64); whereas for costs not representing cost-effectiveness 50 51 244 measures, the median cost was USD $4.15 (IQR $1.72-$23.25). Further, diagnosis and 52 53 treatment costs were often lumped together, and so cannot easily be separated for this 54 245 55 56 246 analysis. 57 58 59 60 61 62 63 64 10 65 247 Figure 3 provides a graphic representation of the costs for diagnosis and/or treatment 1 2 248 per STI type. Figure 3 also provides the number of cost value observations and the median 3 4 5 249 cost for each STI type. Except for “General STIs” and syphilis, there was limited variation in 6 7 250 reported costs. For syphilis, some cost values were negative in that they represented cost 8 9 10 251 savings resulting from treating the infection and eliminating or reducing future expenses as a 11 12 252 result. 13 14 15 253 16 17 254 3.3 Regression analysis 18 19 255 Given the diverse nature of the data and the means with which it was collected and 20 21 22 256 presented, the regression results in Table 2 should be interpreted cautiously. 23 24 257 The results of the two analyses conducted separately with the entire sample of cost 25 26 27 258 values and the sample containing only unit cost values were similar. Analysis with the entire 28 29 259 sample indicated that cost-effectiveness measures were significantly higher than unit cost 30 31 32 260 values. In all three regressions, costs presented as full costs were significantly higher than 33 34 261 incremental costs. In the regression with unit costs only and with all cost values, the cost 35 36 262 values for mobile outreach services were less costly than at fixed clinics. In contrast, in the 37 38 39 263 regression with cost-effectiveness measures only, mobile services were more costly. Also, in 40 41 264 the regression with cost-effectiveness measures only, cost-effectiveness measures were 42 43 44 265 significantly lower in Sub-Saharan Africa than for non-Sub-Saharan African LMICs. 45 46 266 The multiplication factors presented along with the regression coefficients for each 47 48 49 267 variable in the table can shed light on the magnitude of the contribution of each variable to 50 51 268 changes in the cost values. For example, considering the entire sample of costs, costs 52 53 presented as full costs were 41.35 times larger than incremental costs. 54 269 55 56 270 57 58 271 3.4 Cost drivers and HIV 59 60 61 62 63 64 11 65 272 Thirty-seven (84.1%) of the papers qualitatively discussed “cost drivers,” or those 1 2 273 parameters which mattered most in determining the costs of STI diagnosis and treatment. 3 4 5 274 Figure 4 provides a summary of the 141 cost drivers reported in those papers. The type of 6 7 275 intervention or service delivery (e.g. syndromic management, fixed clinic, mobile service, 8 9 10 276 etc.) was most frequently mentioned. 11 12 277 HIV was specifically mentioned in 35 of the 44 papers, as a cost driver (n=7) or in the 13 14 general discussion of STI management. Eleven of the papers discussing HIV included 15 278 16 17 279 evaluations of services which provided integrated, HIV and STI services. 18 19 280 20 21 281 4. Discussion 22 23 24 282 This review summarizes recent literature documenting the costs of diagnosing and 25 26 283 treating STIs in LMICs. Cost figures resulting from cost-effectiveness analyses were nearly 27 28 29 284 14 times higher than unit costs. Unfortunately when examining cost drivers for cost 30 31 285 effectiveness outcomes and unit costs separately, the information presented in the review 32 33 34 286 articles was lacking. Similar to results reported by Terris-Prestholt et al (2006) [11], many of 35 36 287 the articles excluded details on the intervention, including the number of individuals (or 37 38 288 specifically the number of men or women) screened or treated. There was also limited 39 40 41 289 information on how costing was performed, which specific inputs were included or excluded, 42 43 290 and how overhead and infrastructure were treated. In some of the articles, the year the costing 44 45 46 291 was performed was also missing. 47 48 292 Considering the focus of the articles – screening and treating STIs – it was often 49 50 51 293 unclear what kinds of diagnostics were used, and often screening and treatment costs were 52 53 294 bundled together. Finally, frequently the articles presented only general information on the 54 55 295 types of STIs or STI-symptoms that were treated, making it difficult to know how 56 57 58 296 intervention costs should be attributed to specific conditions. 59 60 61 62 63 64 12 65 297 Fortunately, many of the authors provided their own qualitative assessment of the 1 2 298 cost drivers. STI incidence or prevalence, the type and scale of intervention or service 3 4 5 299 delivery, and costs related to the diagnostic equipment and medications were commonly 6 7 300 cited. In the future, it would be helpful if data on these parameters could be published 8 9 10 301 alongside the costs of interventions in these settings. 11 12 302 The frequency of publication of costs for STI screening and treatment seems to have 13 14 increased over time. In late 2005, Terris-Prestholt et al (2006) identified 53 original studies 15 303 16 17 304 spanning a period of 25 years (1980-2005). “Venereal disease” was most commonly 18 19 305 mentioned (28.2% of drug costs and 22.4% of treatment costs) [11]. In contrast, this updated 20 21 22 306 review identified 44 original studies for a period of 8 years (2006-2014); over a third of the 23 24 307 cost values represent screening and treatment of syphilis. 25 26 27 308 Also in this review, there was little variation in reported costs when considering unit 28 29 309 costs only; most were relatively low. Terris-Prestholt (2006) found a high median STI 30 31 32 310 treatment cost of $17.80 (2004 US Dollars) per case as compared to $4.15 (2015 US Dollars) 33 34 311 for diagnostic and/or treatment unit costs in this review. Terris-Prestholt et al noted that 35 36 312 syndromic management protocols were significantly less costly than test-and-treat 37 38 39 313 approaches. In this review, syndromic management was less costly, although the coefficients 40 41 314 were not statistically significant. Contrary to Terris-Prestholt (2006) [11], in this review we 42 43 44 315 found conflicting results for costs in fixed facilities versus mobile services. Mobile services 45 46 316 were less costly when comparing all cost values and unit costs only, but not when comparing 47 48 49 317 cost-effectiveness values. Finally, one surprising finding from this review was that economic 50 51 318 costs appeared lower than financial costs. This was potentially due to the limited number of 52 53 economic costs in the dataset. 54 319 55 56 320 During screening for this review, it was noted that many publications mentioned the 57 58 321 cost effectiveness of screening and treatment for STIs without actually presenting cost 59 60 61 62 63 64 13 65 322 figures. Perhaps due to the relatively low cost of STI screening and treatment, insufficient 1 2 323 attention is given to factors impacting on cost, such as prevalence (including asymptomatic 3 4 5 324 disease), new medications, and new technologies for screening and treatment. Emerging 6 7 325 issues such as point-of-care diagnostics and multi-drug resistance, as has emerged with 8 9 10 326 gonorrhea [59], may prompt renewed interest. 11 12 327 STI incidence and prevalence are significant public health concerns globally. More 13 14 data are needed on specific approaches used to screen and treat for STIs in LMICs in order to 15 328 16 17 329 appropriately plan for service delivery today and the impact of modifying approaches in the 18 19 330 future in line with WHO recommendations. Given the multifaceted relationship between HIV 20 21 22 331 and STI infection and the present emphasis on integrated health care, more information is 23 24 332 also needed on the challenges, benefits, and costs of offering HIV- and STI-targeted services 25 26 27 333 together. 28 29 334 30 31 32 335 Compliance with Ethical Standards 33 34 336 We declare that we have no competing interests related to the research or its potential 35 36 337 application. This work did not involve human subjects and was thus exempt from ethical 37 38 39 338 review and consent requirements. All data are secondary and drawn from published sources. 40 41 339 42 43 340 Acknowledgments 44 45 46 341 The authors gratefully acknowledge contributions from Dr. Fern Terris-Prestholt 47 48 342 during the preparation of this manuscript. 49 50 51 343 This study was made possible by the generous support of the American people 52 53 344 through the United States Agency for International Development (USAID), award number 54 55 345 AID-674-A-12-00029. The contents are the responsibility of the Health Economics and 56 57 58 59 60 61 62 63 64 14 65 346 Epidemiology Research Office, a Division of the Wits Health Consortium (Pty) Ltd and do 1 2 347 not necessarily reflect the views of USAID or the United States Government. 3 4 5 348 6 7 349 References 8 9 10 350 1. World Health Organization. Sexually transmitted infections: Fact sheet N°110 [Internet]. 11 12 351 [cited 2016 Mar 30]. Available from: http://www.who.int/mediacentre/factsheets/fs110/en/ 13 14 15 352 2. World Health Organization. Global incidence and prevalence of selected curable sexually 16 17 18 353 transmitted infections - 2008. Geneva; 2012. 19 20 21 354 3. Looker KJ, Garnett GP, Schmid GP. An estimate of the global prevalence and incidence of 22 23 355 herpes simplex virus type 2 infection. Bull World Health Organ [Internet]. 2008;86:737–816. 24 25 26 356 Available from: http://www.who.int/bulletin/volumes/86/10/07-046128.pdf 27 28 29 357 4. United Nations Department of Economic and Social Affairs. Sustainable Development 30 31 358 Goals [Internet]. Sustain. Dev. Knowl. Platf. 2015 [cited 2015 Nov 26]. Available from: 32 33 34 359 https://sustainabledevelopment.un.org/topics 35 36 37 360 5. Gottlieb SL, Low N, Newman LM, Bolan G, Kamb M, Broutet N. Toward global 38 39 361 prevention of sexually transmitted infections (STIs): The need for STI vaccines. Vaccine. 40 41 42 362 2014;32:1527–35. 43 44 45 363 6. World Health Organization. Guidelines for the Management of Sexually Transmitted 46 47 364 Infections. Geneva; 2003. 48 49 50 365 7. World Health Organization Department of Reproductive Health and Research. Global 51 52 53 366 Health Sector Strategy on Sexually Transmitted Infections 2016-2021: Towards ending STIs. 54 55 367 Geneva; 2016. 56 57 58 368 8. Cepheid. Xpert CT/NG Assay: Product insert. 2012. 59 60 61 62 63 64 15 65 369 9. Cepheid. Xpert HPV [Internet]. [cited 2015 Sep 29]. Available from: 1 2 370 http://www.cepheid.com/en/cepheid-solutions-uk/clinical-ivd-tests/sexual-health/xpert-hpv 3 4 5 371 10. Herbst de Cortina S, Bristow CC, Davey DJ, Klausner JD. A Systematic Review of Point 6 7 8 372 of Care Testing for Chlamydia trachomatis , Neisseria gonorrhoeae , and Trichomonas 9 10 373 vaginalis. Infect Dis Obstet Gynecol [Internet]. 2016; Available from: 11 12 13 374 http://www.hindawi.com/journals/idog/2016/4386127/ 14 15 16 375 11. Terris-Prestholt F, Vyas S, Kumaranayake L, Mayaud P, Watts C. The costs of treating 17 18 376 curable sexually transmitted infections in low-and middle-income countries: A systematic 19 20 21 377 review. Sex Transm Dis [Internet]. 2006;33. Available from: 22 23 378 http://cat.inist.fr/?aModele=afficheN&cpsidt=18210008 24 25 26 379 12. World Bank. Country and Lending Groups [Internet]. 2015 [cited 2017 Feb 11]. 27 28 29 380 Available from: https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world- 30 31 381 bank-country-and-lending-groups 32 33 34 382 13. World Bank. Official exchange rate (LCU per US$, period average) [Internet]. World 35 36 37 383 Bank Data Indic. [cited 2017 Jun 1]. Available from: 38 39 384 data.worldbank.org/indicator/PA.NUS.FCR 40 41 42 385 14. International Monetary Fund. Consumer Price Index [Internet]. World Econ. Outlook 43 44 45 386 Database. [cited 2015 Apr 9]. Available from: 46 47 387 https://www.imf.org/external/pubs/ft/weo/2014/02/weodata/index.aspx 48 49 50 388 15. Colvin M, Bachmann MO, Homan RK, Nsibande D, Nkwanyana NM, Connolly C, et al. 51 52 53 389 Effectiveness and cost effectiveness of syndromic sexually transmitted infection packages in 54 55 390 South African primary care: cluster randomised trial. Sex Transm Infect [Internet]. 56 57 58 391 2006;82:290–4. Available from: http://files/865/PMC2564711.html 59 60 61 62 63 64 16 65 392 16. Legoff J, Bouhlal H, Gresenguiet G, Weiss H, Khonde N, Hocini H, et al. Real-Time 1 2 393 PCR Quantification of Genital Shedding of Herpes Simplex Virus (HSV) and Human 3 4 5 394 Immunodeficiency Virus (HIV) in Women Coinfected with HSV and HIV. J Clin Microbiol. 6 7 395 2006;44:423–32. 8 9 10 396 17. Price MA, Stewart SR, Miller WC, Behets F, Dow WH, Martinson FEA, et al. The cost- 11 12 13 397 effectiveness of treating male trichomoniasis to avert HIV transmission in men seeking 14 15 398 sexually transmitted disease care in Malawi. J Acquir Immune Defic Syndr [Internet]. 16 17 18 399 2006;43:202–9. Available from: http://www.ncbi.nlm.nih.gov/pubmed/16951650 19 20 21 400 18. Vickerman P, Watts C, Peeling RW, Mabey D, Alary M. Modelling the cost effectiveness 22 23 401 of rapid point of care diagnostic tests for the control of HIV and other sexually transmitted 24 25 26 402 infections among female sex workers. Sex Transm Infect [Internet]. 2006 [cited 2013 Aug 27 28 403 15];82:403–12. Available from: 29 30 31 404 http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2563843&tool=pmcentrez&rende 32 33 405 rtype=abstract 34 35 36 406 19. Vickerman P, Terris-Prestholt F, Delany S, Kumaranayake L, Rees H, Watts C. Are 37 38 39 407 targeted HIV prevention activities cost-effective in high prevalence settings? Results from a 40 41 408 sexually transmitted infection treatment project for sex workers in Johannesburg, South 42 43 409 Africa. Sex Transm Dis [Internet]. 2006 [cited 2013 Aug 15];33:S122-32. Available from: 44 45 46 410 http://www.ncbi.nlm.nih.gov/pubmed/16735954 47 48 49 411 20. Vickerman P, Peeling RW, Terris-Prestholt F, Changalucha J, Mabey D, Watson-Jones 50 51 412 D, et al. Modelling the cost-effectiveness of introducing rapid syphilis tests into an antenatal 52 53 54 413 syphilis screening programme in Mwanza, Tanzania. Sex Transm Infect [Internet]. 2006;82 55 56 414 Suppl 5:v38-43. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17215276 57 58 59 415 21. Blandford JM, Gift TL, Vasaikar S, Mwesigwa-Kayongo D, Dlali P, Bronzan RN. Cost- 60 61 62 63 64 17 65 416 effectiveness of on-site antenatal screening to prevent congenital syphilis in rural eastern 1 2 417 Cape Province, Republic of South Africa. Sex Transm Dis [Internet]. 2007;34:S61-6. 3 4 5 418 Available from: http://www.ncbi.nlm.nih.gov/pubmed/17308502 6 7 8 419 22. McClamroch K, Behets F, Van Damme K, Rabenja LN, Myers E. Cost-effectiveness of 9 10 420 treatment strategies for cervical infection among women at high risk in Madagascar. Sex 11 12 13 421 Transm Dis [Internet]. 2007;34:631–7. Available from: 14 15 422 http://www.ncbi.nlm.nih.gov/pubmed/17308503 16 17 18 423 23. Orach CG, Dubourg D, De Brouwere V. Costs and coverage of reproductive health 19 20 21 424 interventions in three rural refugee-affected districts, Uganda. Trop Med Int Heal. 22 23 425 2007;12:459–69. 24 25 26 426 24. Romoren M, Hussein F, Steen TW, Velauthapillai M, Sundby J, Hjortdahl P, et al. Costs 27 28 29 427 and health consequences of chlamydia management strategies among pregnant women in 30 31 428 sub-Saharan Africa. Sex Transm Infect [Internet]. 2007;83:558–66. Available from: 32 33 34 429 http://files/886/PMC2598644.html 35 36 37 430 25. Bukar M, Audu BM, Takai UI, Ajayi BB, Kullima AA. Is routine antenatal screening for 38 39 431 syphilis in Nigeria still justified clinically and economically? Saudi Med J [Internet]. 40 41 42 432 2009;30:1311–5. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19838440 43 44 45 433 26. Corbell C, Stergachis A, Ndowa F, Ndase P, Barnes L, Celum C. Genital Ulcer Disease 46 47 434 Treatment Policies and Access to Acyclovir in Eight Sub-Saharan African Countries. Sex 48 49 50 435 Transm Dis. 2010;37:488–93. 51 52 53 436 27. Suleiman IA, Tayo F. Evaluation of the cost of therapy for the treatment of sexually 54 55 437 transmitted infections in a Nigerian teaching hospital. J Pharm Heal Serv Res [Internet]. 56 57 58 438 2012;3:115–20. Available from: http://onlinelibrary.wiley.com/doi/10.1111/j.1759- 59 60 439 8893.2011.00067.x/full 61 62 63 64 18 65 440 28. Suleiman IA, Tayo F. Comparative costs of antibacterial usage in sexually transmitted 1 2 441 infections in a Nigerian teaching hospital. Trop J Pharm Res [Internet]. 2010;9. Available 3 4 5 442 from: http://www.ajol.info/index.php/tjpr/article/view/63553 6 7 8 443 29. Dize L, West S, Quinn TC, Gaydos CA. Pooling ocular swab specimens from Tanzania 9 10 444 for testing by Roche Amplicor and Aptima Combo 2 assays for the detection of Chlamydia 11 12 13 445 trachomatis: accuracy and cost-savings. Diagn Microbiol Infect Dis [Internet]. 2013;77:289– 14 15 446 91. Available from: http://www.ncbi.nlm.nih.gov/pubmed/24079951 16 17 18 447 30. Kacker S, Frick KD, Quinn TC, Gray RH, Tobian AAR. Financial implications of male 19 20 21 448 circumcision scale-up for the prevention of HIV and other sexually transmitted infections in 22 23 449 Sub-Saharan Africa. Sex Transm Dis. 2013;40:559–68. 24 25 26 450 31. Semini I, Batona G, Lafrance C, Kessou L, Gbedji E, Anani H, et al. Implementing for 27 28 29 451 results: program analysis of the HIV/STI interventions for sex workers in Benin. AIDS Care. 30 31 452 2013;25:S30-39. 32 33 34 453 32. Larson B a. Calculating disability-adjusted-life-years lost (DALYs) in discrete-time. Cost 35 36 37 454 Eff Resour Alloc [Internet]. Cost Effectiveness and Resource Allocation; 2013 [cited 2014 38 39 455 Jan 27];11. Available from: 40 41 42 456 http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3776440&tool=pmcentrez&rende 43 44 457 rtype=abstract 45 46 47 458 33. Omoding D, Katawera V, Siedner M, Boum 2nd Y. Evaluation of the SD Bioline 48 49 50 459 HIV/Syphilis Duo assay at a rural health center in Southwestern Uganda. BMC Res Notes 51 52 460 [Internet]. 2014;7:746. Available from: http://www.ncbi.nlm.nih.gov/pubmed/25339379 53 54 55 461 34. Sweeney S, Mosha JF, Terris-Prestholt F, Sollis KA, Kelly H, Changalucha J, et al. The 56 57 58 462 costs of accessible quality assured syphilis diagnostics: informing quality systems for rapid 59 60 463 syphilis tests in a Tanzanian setting. Heal Policy Plan [Internet]. 2014;29:633–41. Available 61 62 63 64 19 65 464 from: http://www.ncbi.nlm.nih.gov/pubmed/23894075 1 2 3 465 35. Rydzak CE, Goldie SJ. Cost-effectiveness of rapid point-of-care prenatal syphilis 4 5 466 screening in sub-Saharan Africa. Sex Transm Dis [Internet]. 2008;35:775–84. Available 6 7 8 467 from: http://www.ncbi.nlm.nih.gov/pubmed/18607319 9 10 11 468 36. Owusu-Edusei Jr. K, Gift TL, Ballard RC. Cost-effectiveness of a dual non- 12 13 469 treponemal/treponemal syphilis point-of-care test to prevent adverse pregnancy outcomes in 14 15 16 470 sub-Saharan Africa. Sex Transm Dis [Internet]. 2011;38:997–1003. Available from: 17 18 471 http://www.ncbi.nlm.nih.gov/pubmed/21992974 19 20 21 37. Kuznik A, Lamorde M, Nyabigambo A, Manabe YC. Antenatal syphilis screening using 22 472 23 24 473 point-of-care testing in Sub-Saharan African countries: a cost-effectiveness analysis. PLoS 25 26 474 Med [Internet]. 2013;10:e1001545. Available from: 27 28 29 475 http://www.ncbi.nlm.nih.gov/pubmed/24223524 30 31 32 476 38. Levin CE, Steele M, Atherly D, Garcia SG, Tinajeros F, Revollo R, et al. Analysis of the 33 34 477 operational costs of using rapid syphilis tests for the detection of maternal syphilis in Bolivia 35 36 37 478 and Mozambique. Sex Transm Dis [Internet]. 2007;34:S47-54. Available from: 38 39 479 http://www.ncbi.nlm.nih.gov/pubmed/17220812 40 41 42 480 39. White RG, Orroth KK, Glynn JR, Freeman EE, Bakker R, Habbema JDF, et al. Treating 43 44 45 481 curable sexually transmitted infections to prevent HIV in Africa: still an effective control 46 47 482 strategy? J Acquir Immune Defic Syndr [Internet]. 2008;47:346–53. Available from: 48 49 50 483 http://www.ncbi.nlm.nih.gov/pubmed/18176323 51 52 53 484 40. Schackman BR, Neukermans CP, Fontain SN, Nolte C, Joseph P, Pape JW, et al. Cost- 54 55 485 effectiveness of rapid syphilis screening in prenatal HIV testing programs in Haiti. PLoS Med 56 57 58 486 [Internet]. 2007;4:e183. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17535105 59 60 61 62 63 64 20 65 487 41. Benzaken AS, Sabido M, Galban EG, Pedroza V, Vasquez F, Araujo A, et al. Field 1 2 488 evaluation of the performance and testing costs of a rapid point-of-care test for syphilis in a 3 4 5 489 red-light district of Manaus, Brazil. Sex Transm Infect [Internet]. 2008;84:297–302. 6 7 490 Available from: http://www.ncbi.nlm.nih.gov/pubmed/18305119 8 9 10 491 42. Aldridge RW, Iglesias D, Caceres CF, Jaime Miranda J. Determining a cost effective 11 12 13 492 intervention response to HIV/AIDS in Peru. BMC Public Health [Internet]. 2009;9. Available 14 15 493 from: http://www.biomedcentral.com/1471-2458/9/352/ 16 17 18 494 43. Alarid-Escudero F, Sosa-Rubí SG, Fernández B, Galárraga O. [Cost-benefit analysis: 19 20 21 495 HIV/AIDS prevention in migrants in Central America]. Salud Publica Mex [Internet]. 22 23 496 2013;55 Suppl 1:S23-30. Available from: http://www.ncbi.nlm.nih.gov/pubmed/23918053 24 25 26 497 44. Tan DHS, Raboud JM, Kaul R, Grinsztejn B, Cahn P, Walmsley SL. Can herpes simplex 27 28 29 498 virus type 2 suppression slow HIV disease progression: a study protocol for the 30 31 499 VALacyclovir in delaying antiretroviral treatment entry (VALIDATE) trial. BiomedCentral 32 33 34 500 [Internet]. BioMed Central Ltd; 2010;11:113. Available from: 35 36 501 http://www.trialsjournal.com/content/11/1/113 37 38 39 502 45. Garcia SG, Tinajeros F, Revollo R, Yam EA, Richmond K, Diaz-Olavarrieta C, et al. 40 41 42 503 Demonstrating public health at work: a demonstration project of congenital syphilis 43 44 504 prevention efforts in Bolivia. Sex Transm Dis [Internet]. 2007;34:S37-41. Available from: 45 46 47 505 http://www.ncbi.nlm.nih.gov/pubmed/17179776 48 49 50 506 46. Chen S, Li J, Van den Hoek A. Universal screening or prophylactic treatment for 51 52 507 Chlamydia trachomatis infection among women seeking induced abortions: Which strategy is 53 54 55 508 more cost-effective? Sex Transm Dis [Internet]. 2007;34:230–6. Available from: http://0- 56 57 509 graphics.tx.ovid.com.innopac.wits.ac.za/ovftpdfs/FPDDNCJCFGPOGK00/fs046/ovft/live/gv 58 59 510 023/00007435/00007435-200704000-00009.pdf 60 61 62 63 64 21 65 511 47. Sharma D, Sethi S, Mehta S Das, Sharma M. In-House Growth-Promoting Transport 1 2 512 System for Neisseria gonorrhoeae ᰔ. J Clin Microbiol. 2007;45:2743–4. 3 4 5 513 48. Abraham A, Babu M, Kavitha S, Jesudason M, Sridharan G. A molecular method for 6 7 8 514 typing Herpes Simplex Virus isolates as an alternative to immunofluorescence methods. 9 10 515 Indian J Med Microbiol. 2009;27:22–6. 11 12 13 516 49. Madhivanan P, Krupp K, Hardin J, Karat C, Klausner JD, Reingold AL. Simple and 14 15 16 517 inexpensive point-of-care tests improve diagnosis of vaginal infections in resource 17 18 518 constrained settings. Trop Med Int Heal [Internet]. 2009;14:703–8. Available from: 19 20 21 519 http://onlinelibrary.wiley.com/store/10.1111/j.1365-3156.2009.02274.x/asset/j.1365- 22 23 520 3156.2009.02274.x.pdf?v=1&t=hvs1qq5d&s=91d0c9198b4d807d3fd1dbf03830a5282bcc82d 24 25 26 521 d 27 28 29 522 50. Dandona L, Kumar SGP, Kumar GA, Dandona R. Cost-effectiveness of HIV prevention 30 31 523 interventions in Andhra Pradesh state of India. BMC Health Serv Res [Internet]. 2010;10. 32 33 34 524 Available from: http://www.ncbi.nlm.nih.gov/pubmed/20459755 35 36 37 525 51. Ha YE, Peck KR, Joo E-J, Kim SW, Jung S-I, Chang HH, et al. Impact of First-Line 38 39 526 Antifungal Agents on the Outcomes and Costs of Candidemia. Antimicrob Agents 40 41 42 527 Chemother. 2012;56:3950–6. 43 44 45 528 52. McCormick DF, Rahman M, Zadrozny S, Alam A, Ashraf L, Neilsen GA, et al. 46 47 529 Prevention and control of sexually transmissible infections among hotel-based female sex 48 49 50 530 workers in Dhaka, Bangladesh. Sex Heal [Internet]. 2013;10:478–86. Available from: 51 52 531 http://www.ncbi.nlm.nih.gov/pubmed/24262217 53 54 55 532 53. Owusu-Edusei Jr. K, Tao G, Gift TL, Wang A, Wang L, Tun Y, et al. Cost-effectiveness 56 57 58 533 of integrated routine offering of prenatal HIV and syphilis screening in China. Sex Transm 59 60 534 Dis [Internet]. 2014;41:103–10. Available from: 61 62 63 64 22 65 535 http://www.ncbi.nlm.nih.gov/pubmed/24413489 1 2 3 536 54. Hong FC, Liu JB, Feng TJ, Liu XL, Pan P, Zhou H, et al. Congenital syphilis: an 4 5 537 economic evaluation of a prevention program in China. Sex Transm Dis [Internet]. 6 7 8 538 2010;37:26–31. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19734825 9 10 11 539 55. Kasymova N, Johns B, Sharipova B. The costs of a sexually transmitted infection 12 13 540 outreach and treatment programme targeting most at risk youth in Tajikistan. Cost Eff Resour 14 15 16 541 Alloc C/E [Internet]. 2009;7. Available from: 17 18 542 http://www.ncbi.nlm.nih.gov/pubmed/19883515 19 20 21 56. Cornier N, Petrova E, Cavailler P, Dentcheva R, Terris-Prestholt F, Janin A, et al. 22 543 23 24 544 Optimising the management of vaginal discharge syndrome in Bulgaria: cost effectiveness of 25 26 545 four clinical algorithms with risk assessment. Sex Transm Infect. 2010;86:303–9. 27 28 29 30 546 57. Asicioglu O, Gungorduk K, Ozdemir A, Ertas IE, Yildirim G, Sanci M, et al. Single daily 31 32 547 dose of moxifloxacin versus ofloxacin plus metronidazole as a new treatment approach to 33 34 548 uncomplicated pelvic inflammatory disease: a multicentre prospective randomized trial. Eur J 35 36 37 549 Obs Gynecol Reprod Biol [Internet]. 2013;171:116–21. Available from: 38 39 550 http://www.ncbi.nlm.nih.gov/pubmed/23993130 40 41 42 551 58. Vickerman P, Ndowa F, Mayaud P. Modelling the cost per ulcer treated of incorporating 43 44 45 552 episodic treatment for HSV-2 into the syndromic algorithm for genital ulcer disease. Sex 46 47 553 Transm Infect [Internet]. 2008;84:243–8. Available from: 48 49 50 554 http://www.ncbi.nlm.nih.gov/pubmed/18385226 51 52 53 555 59. Centers for Disease Control and Prevention (CDC). CDC Grand Rounds: The Growing 54 55 556 Threat of Multidrug-Resistant Gonorrhea [Internet]. Morb. Mortal. Wkly. Rep. 2013 [cited 56 57 58 557 2016 Aug 17]. Available from: 59 60 558 http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6206a3.htm 61 62 63 64 23 65 559 Figure legends 1 560 2 561 Figure 1: Search and screening results 3 562 1 This included the original systematic review we are updating (Terris-Prestholt et al, 2006), and an unpublished 4 563 PhD thesis (we included published literature only). 5 6 564 7 565 8 566 Figure 2: Frequency of categorization of costs (N=202) 9 567 10 568 11 12 569 Figure 3: Unit costs for diagnosis and/or treatment of STIs by STI type (n=202), USD 13 570 $2015 14 571 CT = chlamydia; NG=gonorrhea; CT & NG = Costs presented for chlamydia and gonorrhea treatment jointly; 15 572 TV = trichomoniasis; TV & BV = Costs presented for trichomoniasis and bacterial vaginosis jointly; PID = 16 573 Pelvic inflammatory disease; GUD = Genital ulcerative disease; Syphilis & GUD = Costs presented for syphilis 17 574 and GUD jointly; TV, BV, GUD = Costs presented for trichomoniasis, bacterial vaginosis, and GUD jointly; 18 575 Other = Other, non-specified STIs 19 576 *NB: Data represent a mixture of outcomes, e.g. cost per person treated, per person treated successfully, per 20 577 HIV infection averted, etc. Box plots represent median and IQR. 21 578 22 579 23 Figure 4: Cost drivers (n=141) as reported in papers discussing factors affecting costs 24 580 25 581 (n=37) 26 582 27 583 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 24 65 Table Click here to download Table Lince-Deroche. STI Dx and Tx Cost - Table 1 2018.02.12.docx

Table 1: Selected literature by year, summary of STI screening and treatment interventions and unit costs/cost effectiveness measures Country of Study Description Target Type of Costing Intervention description Outcomes (per) Unit Cost/Cost Study, Population, STI/ method, Effectiveness Author, Year Sites, N approach perspective, (2013 $US) and (Minimum, approach Maximum) South Africa, Cost effectiveness of General, SynM Emp/Inc/Eco/ Pre-packed medication for Extra patient 1.71 Colvin, syndromic management 37 sites, Prov/Bot syndromic treatment of STIs correctly treated 200615 in cluster randomized 256 during PHC-based Extra patient 127.63 trial at primary health individuals consultations appropriately care (PHC) facilities managed Ghana, Evaluation of the Women co- HSV2 Emp/Inc/Fin/ Real-time PCR on a Light Cost per reaction 10.18 Legoff, accuracy and usefulness infected Prov/Bot Cycler instrument 16 2006 of laboratory-developed with HIV Branched DNA assay Price of HIV-1 56.05 real-time PCR procedures and HSV2, 2 RNA quantification for women co-infected sites, -- with HIV and HSV2 Use of MagNA Pure system for Cost per test 4.99 combined extraction of HIV and HSV genomes from genital excretions prior to PCR Use of manual extraction by Cost per test 3.69 High Pure viral nucleic acid kit for combined extraction of HIV and HSV genomes from genital excretions prior to PCR Malawi, Cost effectiveness of Male STI TV Mod/Inc/Fin/ Syndromic management for Case treated 0.06 Price, 200617 treating male TV clinic Prov/Ing men with urethritis HIV infection 29.95 attendees, --, averted 10,000 Syndromic management for Case treated 0.06 individuals men with urethritis and GUD HIV infection 14.71 averted Syndromic management for all Case treated 0.07 men HIV infection 14.32 averted

Status quo (All male Case treated 0.03 asymptomatic partners of symptomatic women who receive metronidazole are treated with 2g of metronidazole) South Africa, Cost effectiveness of FSWs, SynM Emp/Ful/Eco FSW mobile STI clinic Woman seen 114.39 Vickerman, syndromic management, 1 site, (PPT) /Prov/Ing Woman seen 200.56 200619 condoms, and periodic 1,431 repeatedly presumptive treatment individuals Visit 65.37 (PPT) of female sex Syndrome treated 151.53 workers (FSWs) Emp/Ful/Fin/ FSW mobile STI clinic Woman seen 103.99 Prov/Ing Woman seen 184.22 repeatedly Visit 59.42 Syndrome treated 138.16 Model Mod/Ful/Eco Syndromic treatment only HIV infection 2,837.52 (1,834.73, assumed /Prov/Ing averted 5,054.05) treatment Condom, syndromic treatment, HIV infection 3,109.39 (2,056.09, of CT, presumptive treatment averted 5,400.20) NG, TP, Syndromic treatment only DALY averted 105.48 (69.82, chanc., 167.87) GUD Condom, syndromic treatment, DALY averted 115.88 (78.74, presumptive treatment 179.76) Benin, Incremental cost FSWs, CT, NG Mod/Inc/Fin/ POC testing for CT and NG HIV infection (128.74, 181.68) Vickerman, effectiveness of rapid 1 site, -- Prov/Ing averted 200618 point-of-care (POC) Mod/Inc/Eco/ POC testing for CT and NG Per test 1.20, 4.81) testing for FSWs Prov/Ing Per additional HIV (140.17, 294.66) infection averted Tanzania, Modelled cost Antenatal TP Mod/Ful/Fin/ Syphilis screening Per DALY averted (14.59, 42.94) Vickerman, effectiveness analysis of clinic, --, -- Prov/Ing 200620 rapid TP testing South Africa, Cost effectiveness Antenatal TP Emp/Inc/Fin/ On-site RPR Per mother- 4.92 Blandford, analysis of antenatal clinic Prov/Bot Off-site RPR/TPHA newborn pair tested 4.73 200721 screening for TP patients, 2 On-site ICS 6.30 sites, 1,000 Mod/Inc/Fin/ Off-site RPR/TPHA Per case averted 134.96 individuals Prov/Bot On-site ICS 149.34

China, Chen, Modelled cost Women CT Mod/Unk/Fin Azithromycin-based Case of PID 90.64 200746 effectiveness of universal seeking /Prov/Ing prophylaxis prevented screening and induced Universal screening by PCR Case of PID 702.02 prophylactic treatment for abortion, 2 prevented CT sites, 10,000 individuals Bolivia, Demonstration project of Antenatal TP Emp/Inc/Fin/ POC TP testing Woman tested 1.83 Garcia, introducing POC clinic Prov/Ing 2007*45 antenatal TP testing patients, 41 sites, 11,618 individuals Bolivia, Operational costs analysis Bolivian TP Emp/Inc/Eco/ POC TP testing (RPR) Woman screened 3.11 Mozambique, of POC TP testing antenatal Prov/Bot Levin, 200738 clinic POC TP testing (ICS) Women screened 4.15 patients, 41 sites, 11,618 POC TP testing (ICS, rural Woman screened 6.17 individuals health facilities) POC TP testing and treatment Woman screened 4.78 (RPR) and treated POC TP testing and treatment Woman screened 5.54 (ICS) and treated POC TP testing and treatment Woman screened 5.41 (ICS, rural health facilities) and treated Mozambica TP Emp/Inc/Eco/ POC TP testing (RPR) Woman screened 1.06 n antenatal Prov/Bot POC TP testing (ICS) Woman screened 1.22 clinic POC TP testing (ICS, rural Woman screened 1.19 patients, 164 health facilities) sites, 33,097 POC TP testing and treatment Woman screened 1.94 individuals (RPR) and treated POC TP testing and treatment Woman screened 1.98 (ICS) and treated POC TP testing and treatment Woman screened 1.95 (ICS, rural health facilities) and treated Madagascar, Cost effectiveness of Female sex Gen Mod/Inc/Eco/ (First visit) Presumptive Woman treated 6.86 McClamroch, treatment strategies for workers, --, Soc/Ing treatment/(Follow-up visit) Prevented case of 48.68 200722 cervical infection -- Risk-based treatment (National PID considering changes in Guidelines)

national policy from first (First visit) Risk-based Woman treated 6.92 to follow-up visit treatment/ /(Follow-up visit) Prevented case of 45.95 Presumptive treatment PID (First visit) Presumptive Woman treated 7.06 treatment//(Follow-up visit) Prevented case of 47.95 Presumptive treatment PID Uganda, Costs and coverage of Refugee Gen Emp/Inc/Fin/ Syndromic management in Service provided 2.21 Orach, reproductive health affected Prov/Ing hospitals (for refugee hosts) 2007*23 interventions districts, 38 Syndromic management in HC Service provided 1.73 sites, 585 IV (refugee hosts) individuals Syndromic management in HC Service provided 1.36 III (refugees) 1.67 Syndromic management in HC Service provided 1.42 III (refugee hosts) 1.74 Syndromic management in HC Service provided 1.20 II (refugees) 1.47 Syndromic management in HC Service provided 1.39 II (refugee hosts) 1.70 TP Emp/Inc/Fin/ TP treatment in hospital Service provided 3.70 Prov/Ing (refugee hosts) TP treatment in HC IV (refugee Service provided 2.99 hosts) TP treatment in HC III Service provided 3.05 (refugees) TP treatment in HC III (refugee Service provided 3.58 hosts) Botswana, Decision analysis model Antenatal CT Mod/Inc/Fin/ Erythromycin treatment Episode 1.86 (0.52, 7.44) Romoren, for costs and care Prov/Ing Azithromycin treatment Episode 1.05 (0.59, 2.16) 200724 consequences of CT attendees, --, Doxycycline treatment Episode 0.53 (0.25, 2.06) management 100,000 POC testing Episode 1.09 (0.62, 4.54) individuals Syndromic management with Cured case 70.47 erythromycin treatment Syndromic management with Cured case 22.42 azithromycin treatment POC testing and treatment Cured case 33.10 Haiti, Cost-effectiveness Antenatal TP Mod/Inc/Fin/ Syndromic surveillance Patient tested 0.89 Schackman, analysis of rapid TP clinic Soc/Bot TP screening (RPR, rural) Patient tested 2.65 200740 TP screening (Rapid, rural) Patient tested 3.98

testing with immediate patients, --, - TP screening (RPR, urban) Patient tested 10.21 treatment - TP screening (Rapid, urban) Patient tested 12.89 TP screening (Rapid, rural) DALY averted 12.65 TP screening (Rapid, urban) DALY averted 18.43 India, In-house transport system Men with NG Emp/Inc/Fin/ Use of Eno transport system for Cost per sachet 0.05 Sharma, for Neisseria gonorrhoeae symptoms of Prov/Bot NG 200747 samples acute urethritis, 2 sites, -- Brazil, Field evaluation of POC STI clinic TP Emp/Inc/Fin/ Laboratory based TP testing Case of syphilis 18.34 Benzaken, TP test attendees, 1 Prov/Bot POC TP screening Case of syphilis (36.24, 61.45) 200841 site, 510 Laboratory based TP testing Case of active 23.25 individuals syphilis POC TP screening Case of active (62.76, 106.53) syphilis Sub-Saharan Modelled cost Antenatal TP Mod/Inc/Fin/ POC TP testing (ICS) Lifetime cost 213.31# Africa, effectiveness of POC TP clinic Soc/Ing savings per woman Rydzak, testing patients, --, POC TP testing (RPR) Lifetime cost 202.11 200835 1,000 savings per woman individuals TP testing (RPR+TPHA) Lifetime cost 138.25# savings per woman Unspecified, Modelled cost analysis of People HSV2, Mod/Inc/Fin/ Syndromic management of Ulcer treated (<4.25)# Vickerman, syndromic management presenting GUD Prov/Ing GUD 200858 of HSV2 treatment with GUD, - HSV2 treatment Episode 0.64# among patients presenting -, -- with GUD Tanzania, Modelled cost General, 49 Gen Mod/Ful/Fin/ Syndromic management of HIV infection 398.06 (Tanzania) Zambia, effectiveness analysis of sites, 12,537 Prov/Ing STIs averted 682.65 (Zambia) Kenya, syndromic management individuals 3344.94 (Kenya) Cameroon, 2865.59 (Cameroon) Benin, White, 8713.19 (Benin) 200839 India, Cost evaluation of two Tertiary care HSV2 Emp/Ful/Fin/ HSV typing assay (PCR) Cost per PCR 36.30 Abraham, methods for HSV isolate patients, --, Prov/Bot HSV typing assay (MAb IF Cost per MAb IF 42.35 200948 typing 42 assay) assay individuals (50 tests)

Peru, Cost effectiveness MSM, FSW Gen Emp/Inc/Fin/ General STI treatment STI case treated 4.18 Aldridge, analysis of interventions and pregnant Prov/Ing 200942 against HIV/AIDS women, Mod/Inc/Fin/ General STI treatment (50% DALY averted 835.26 (226.05, transmission 1,000 Prov/Ing coverage) 1,154.00) individuals General STI treatment (80% DALY averted 857.87 (224.92, coverage) 1,181.12) General STI treatment (95% DALY averted 866.91 (226.05, coverage) 1,192.43) Nigeria, Seroprevalence and cost Antenatal TP Emp/Inc/Fin/ Antenatal TP screening Woman tested 2.80 Bukar, effectiveness analysis of clinic Prov/Bot 2009*25 antenatal TP screening patients, 1 site, 18,712 individuals Tajikistan, Costing of STI outreach Most at risk NG, Emp/Inc/Fin/ STI outreach and screening test NG test 3.42 Kasymova, programme adolescents TP, Prov/Bot TP test 1.67 200955 (sex HSV2, HSV2 test 3.72 workers, CT, CT test 3.42 IDU, MSM, TV, TV test 3.42 street Gen Average for any 2.86 children), 8 STI test sites, 8,020 STI outreach and treatment NG case treated 8.18 individuals TP case treated 3.02 HSV2 case treated 1.24 CT case treated 7.94 TV case treated 0.33 Average for any 4.18 STI case treated STI outreach, screening and Adolescent tested 18.26 treatment and treated India, Effectiveness evaluation Sexually TV, BV Emp/Inc/Fin/ POC testing for BV/TV Woman tested 0.05 Madhivanan, of POC testing for active Prov/Ing 2009*49 vaginal infections women attending reproductive health clinics, 2 sites, 898 individuals

Botswana, Scoping study evaluating Public GUD Emp/Inc/Fin/ Acyclovir (200 mg) Cost per tablet 3.52 (Botswana) Kenya, uptake and use of health Prov/Bot Cost per tablet 1.50 (South Africa) Malawi, acyclovir for GUD facility, --,-- Cost per tablet 1.49 (Uganda) South Africa, treatment Cost per tablet 3.30 (Tanzania) Tanzania, Cost per tablet 1.72 (Kenya) Uganda, Acyclovir (400 mg) Cost per tablet 1.32 (Botswana) Corbell, Cost per tablet 0.94 (Malawi) 26 2010 Cost per tablet 1.32 (South Africa) Bulgaria, Cost effectiveness of risk- Non- CT, NG Emp/Inc/Fin/ Risk-score adapted syndromic Patient treated 3.48 Cornier, score adjusted syndromic pregnant Prov/Bot management (Algorithm 1) true case treated 43.42 201056 management women Risk-score adapted syndromic Patient treated 1.68 presenting management (Algorithm 2) true case treated 37.09 with VDS at Risk-score adapted syndromic Patient treated 3.62 an NGO management (Algorithm 3) true case treated 43.16 sexual Risk-score adapted syndromic Patient treated 6.40 health clinic, management (Algorithm 4) true case treated 69.57 1 site, 424 individuals India, Cost effectiveness of HIV STI clinic Gen Emp/Ful/Fin/ STI clinics Person served 23.74 Dandona, prevention interventions attendees, --, Prov/Bot HIV infection 1196.95 (904.23, 201050 22 sites averted 1,624.43) DALY saved 43.47 (31.88, 57.96.23) DALY saved (non- 44.92 (34.78,62.31 age weighted) China, Hong, Economic evaluation of Antenatal TP Emp/Inc/Fin/ TP screening Woman screened 6.99 201054 TP screening clinic Soc/Bot Infection identified 1,344.80 patients, 61 sites, Case of syphilis 7,668.87 159,017 averted individuals Case of low birth 8,968.27 weight averted Death averted 12,356.47 DALY saved 375.50 Nigeria, Evaluation of the cost of Tertiary Gen Emp/Inc/Fin/ Antibacterial treatment Patient treated 14.13 Suleiman, antibacterial usage to health Prov/Ing 201028 treat STIs facility (teaching

hospital), 1 site, 230 individuals Argentina, Trial of VALacyclovir for HIV positive HSV2 Emp/Inc/fin/ Use of VALacyclovir to Cost for 163.01 Tan, 201044 delaying need for adults with Prov/Ing suppress HSV-2 VALacyclovir per initiating ART in HIV-1 HSV2, 32 patient per month HSV2 co-infected sites, 480 individuals individuals Sub-Saharan Modelled cost Antenatal TP Mod/Inc/Fin/ POC TP testing Woman tested 4.11 (0.55, 5.51) Africa, effectiveness of POC TP clinic Prov/Ing Owusu- testing patients, --, Mod/Inc/Fin/ POC TP testing and treatment DALY averted 404.64 Edusei, 1,000 Soc/Ing (RPR+TPPA) DALY averted 263.97 2011#36 individuals (including newborn 265.23 outcomes) POC TP testing and treatment DALY averted 372.66 (RPR) 374.43 DALY averted 244.15 (including newborn 245.32 outcomes) POC TP testing and treatment DALY averted 290.90 (Dual-POC) DALY averted 188.67 (including newborn outcomes) POC TP testing and treatment DALY averted 256.68 (ICS) DALY averted 164.58 (including newborn 165.37 outcomes) South Korea, Retrospective cost-of- Tertiary care Candida Emp/Ful/Fin/ First antifungal therapy Cost per patient on 6,053.27 Ha, 201251 illness study focused on patients with Prov/Bot first line therapy candidemia candidemia, First antifungal therapy was Cost per patient on 5,477.68 4 sites, 199 Fluconazole first line therapy individuals First antifungal therapy was Cost per patient on 5,117.78 Amphotericin B deoxycholate first line therapy Nigeria, Evaluation of the cost of Tertiary TP, Emp/Inc/Fin/ Treatment for TP/GUD Patient treated 24.31 (23.11, 25.51) Suleiman, treating STIs in Nigeria health NG, Soc/Bot Treatment for NG Patient treated 23.07 (20.83, 25.30) 201227 facility CT, Treatment for CT Patient treated 66.54 (52.05, 81.03) (teaching Chanc, Treatment for chancroid Patient treated 18.41 (15.72, 21.10) hospital), 1 PID, Treatment for PID Patient treated 21.21 (19.02, 23.40)

site, 227 Gen Treatment for other STIs Patient treated 24.92 (21.03, 28.81) individuals (unspecified) Treatment for any STIs Patient treated 32.12 Guatemala, Cost-benefit analysis of Migrants, 11 Gen Emp/Inc/Fin/ Syndromic management STI case treated 0.79 (Guatemala) Honduras, El HIV prevention for sites, 3,210 Prov/Bot 1.18 (Honduras) Salvador, migrants in Central individuals 0.99 (El Salvador) Nicaragua, America 0.56 (Nicaragua) Panama, 0.98 (Panama) Alarid- Escudero, 201343 Turkey, RCT of treatment Women PID Emp/Inc/Fin/ Moxifloxacin Patient 34. 20 Asicioglu, strategies for diagnosed Prov/Ing Ofloxacin plus metronidazole Patient 25. 49 2013*57 uncomplicated PID with uncomplicat ed PID in hospitals, 4 sites, 1,156 individuals Tanzania, Evaluation of screening --, --, -- CT Emp/Inc/Fin/ Detecting CT by Amplicor CT Sample tested 10.01 Dize, 2013*29 for chlamydia by pooling Unk/Ing specimens Uganda, Markov modelling to 15-49 year GUD, BV, Mod/Inc/Fin/ Medical male circumcision GUD treatment 1.48 Kacker, explore cost effectiveness old men, 1 TV, HIV Prov/Ing costs (with 201330 of male circumcision site, 500 000 acyclovir) of men scale-up (including individuals and women averted impacts on women) Medical male circumcision BV treatment costs 0.47 (with metronidazole and clotrimazole) of women averted Medical male circumcision TV treatment costs 0.36 (with metronidazole) of women averted Medical male circumcision Follow-up visit 9.21 costs for GUD,

BV, and TV averted 23.6% 15-49 year old men MC procedures and 111.27 circumcised over a 5 year MC reduced period infections in an average male MC-reduced 110.19 infections in an average male MC-reduced 20.37 infections in an average female 23.6% 15-49 year old men MC procedures and 15.57 circumcised over a 25 year MC reduced period infections in an average male MC-reduced 15.23 infections in an average male MC-reduced 13.72 infections in an average female Sub-Saharan Cost-effectiveness Antenatal TP Mod/Inc/Fin/ POC TP testing and treatment Screening 1.29 (0.87, 3.32) Africa, analysis of POC TP clinic Prov/Ing Patient treated 6.40 (7.80, 14.76) Kuznik, testing patients, --, - DALY averted 11.59 (5.27, 81.13) 201337 - Bangladesh, Periodic presumptive Hotel-based Gen Emp/Inc/Fin/ Extended syndromic Visit 18.61 McCormick, treatment vs enhanced female sex Prov/Bot management 201352 syndromic management workers, 2 Periodic presumptive treatment Visit 9.29 of STIs sites, 549 individuals Benin, Program analysis of Female sex Gen Emp/Inc/Fin/ Syndromic management of STI Visit 50.54 Semini, HIV/STI interventions for workers, 56 Prov/Top Syndromic management of STI Visit 23.61 (5.80, 85.33) 2013*31 sex workers sites, 3,372 at high volume facilities individuals Syndromic management of STI Visit 399.09 at low volume facilities

Emp/Inc/Fin/ Syndromic management of STI Visit 23.69 Prov/Bot at high volume facilities (Bohicon) Syndromic management of STI Visit 38.86 at high volume facilities (Cotonou) Zambia, Cost and cost- Antenatal TP Emp/Inc/Fin/ POC TP testing Per patient 1.62 Larson, effectiveness of TP clinic Prov/Bot DALY averted 475.53 201432 diagnosis and treatment patients, 18 POC TP testing with HIV test Per patient 2.35 sites, 1,000 individuals Uganda, Field evaluation of POC Antenatal TP Emp/Inc/Fin/ POC TP testing Test kit 1.69 Omoding, TP test clinic Unk/Ing 2014*33 patients, 1 site, 220 individuals China, Modelled cost Antenatal TP Mod/Inc/Fin/ Antenatal TP screening and DALY averted 209.80 Owusu- effectiveness of antenatal clinic Soc/Ing treatment Edusei, TP screening patients, --, Antenatal TP screening and DALY averted 448.317 201453 10,000 treatment and HIV screening individuals Mod/Inc/Fin/ Antenatal TP screening (RPR) Patient tested 3.75 (1.87, 5.62) Prov/Ing Antenatal TP screening Patient test 7.49 (3.75, 11.24) (TPHA) Tanzania, Cost analysis of RST vs Antenatal TP Emp/Inc/Eco/ TP screening (RST) Woman screened 1.84 (1.67, 3.01) Sweeney, RPR TP tests clinic Prov/Bot TP screening (RST) Woman treated 20.52 201434 patients, 9 sites, 6,362 individuals Antenatal TP Emp/Inc/Eco/ TP screening (RPR) Woman screened 2.23 (1.69, 2.90) clinic Prov/Bot TP screening (RPR) Woman treated 12.45 (5.78, 71.36) patients, 9 sites, 224 individuals General: DALY = disability-adjusted life year; FSW = Female sex workers; g = grams; IDU = injecting drug users; MARA = Most at risk adolescents (sex workers, intravenous drug users, men who have sex with men, and street children); MSM = men who have sex with men; NGO = non-governmental organization; PHC = primary health care, POC = point-of-care; RCT = randomized-controlled trial; STI = sexually transmitted infection STI Type, Intervention: BV = Bacterial vaginosis; Chanc = Chancroid; CT = chlamydia (Chlamydia trachomatis); Dual-POC = Dual non-treponemal/treponemal syphilis point-of-care test; Gen = general, unspecified STI; GUD = Genital ulcerative disease; HSV2 = Herpes simplex virus type 2; ICS = Immunochromatographic strip; NG =

gonorrhoea (Neisseria gonorrhoeae); PCR = Polymerase chain reaction; PID = pelvic inflammatory disease; PPT = Periodic presumptive treatment; Rapid = rapid testing; RB = Risk-based treatment; RPR = ; SynM = Syndromic management; Syph = Syphilis; TPHA = T. pallidum hemaglutination assay; TP = syphilis (Treponema pallidum); TV = Trichomonas (Trichomoniasis vaginalis ); VDS = vaginal discharge syndrome Costing method: Emp = Empirical; Mod = Modelled; Ful = Full; Inc = Incremental; Eco = Economic; Fin = Financial; Unk = Unknown Costing perspective: Prov = Provider; Pat = Patient; Soc = Societal Costing approach: Bot = Bottom-up; Top = Top-down; Ing = Ingredients; Unk = Unknown

* Year of study assumed as year in which cost data was presented # Currencies expressed in USD, and US CPI used for inflation adjustment since country of origin was missing.

Figure 1 Click here to download Figure Fig 1.eps Figure 2 Click here to download Figure Fig 2.eps Figure 3 Click here to download Figure Fig 3.eps Figure 4 Click here to download Figure Fig 4.eps Response from STD Click here to download Other Response from STD.pdf 2/12/2018 Gmail - Your Submission

Naomi Lince-Deroche

Your Submission

Sexually Transmitted Diseases Tue, Nov 14, 2017 at 1:40 PM Reply-To: Sexually Transmitted Diseases To: Naomi Lince-Deroche

Ref.: Ms. No. STD17-330 The costs of diagnosing and treating sexually transmitted infections in low- and middle-income countries from 2006 to 2014: An updated systematic review Sexually Transmitted Diseases

Dear Dr Lince-Deroche,

Your manuscript has been reviewed by the reviewers and an Associate Editor. The reviewers had serious reservations about its suitability for publication in Sexually Transmitted Diseases. Please see the comments from the reviewers. Thus, I regret to inform you that the priority assigned to the manuscript was not high enough to allow us to accept it for publication.

Thank you for giving us the opportunity to consider your work.

Yours sincerely,

William C. Miller, MD, PhD, MPH Editor Sexually Transmitted Diseases

Reviewer #1: Summary This study reviewed recent cost-related studies on sexually transmitted infections (STI) from low- and middle income countries to estimate the unit cost of STI diagnosis and treatment.

Major concerns Obviously, it appears that the original article was relevant to some readers, but I have my doubts in regards to the utility of this study.

First, STIs are very different with different natural histories (lifelong vs. short duration). So, to come up with a single unit cost encompassing ALL will be very difficult and, in my opinion, should not be done. If there was/were any rationale for coming up with ONE unit cost, then that was not presented, or it is unclear.

Second, given the huge disparity in the magnitude of the costs, a single unit cost that does not account for the distribution (or is not weighted) or relative burden of these STIs can be very deceptive.

I have been mulling over this question ever since I started reading this manuscript--can the authors specify/clarify how this estimate will be used?

Reviewer #2: OVERALL This systematic review summarises the literature on an important issue for STI programme design/priority setting - the current literature on the costs of diagnosis and treatment of curable STIs in a low and middle income settings. The systematic search and analysis replicate (and update) an earlier review by Terris-Prestholt et al (2006). The systematic search is comprehensive and the statistical analysis is based on earlier published work. Overall, it is my impression that there are three main areas which could improve the manuscript: (1) the methods section would benefit from additional detail/definitions of cost data that are needed to understand the analysis undertaken and presented; (2) I have a concern about the heterogeneity of the information included within a single regression model, specifically not all of the variables that describe the cost data are included as explanatory variables in the model (e.g. specific STI; provider or societal perspective); (3) in my opinion, a key strength of this paper is in the collection/summary of the available data on the costs of diagnosis and treating STIs in LMICs rather than the statistical combination of these findings. I think the manuscript would be improved by more consideration of the available published cost data (line 231 onwards). And more detailed presentation or analysis of the cost drivers.

https://mail.google.com/mail/u/0/?ui=2&ik=0f32d53aae&jsver=FOnR4BGjAPw.en.&view=pt&msg=15fba55100fe25e6&q=csfirnhaber%40gmail.com&qs… 1/3 2/12/2018 Gmail - Your Submission

ABSTRACT 1. The abstract does not introduce the reader to what is already known in this area or why this study is important. 2. Similarly the conclusion reflects on the lack of available information and does not include a reflection of the information presented in the results section. It is my opinion that if the authors consider there to be sufficient available information to undertake the planned statistical analysis, then the key findings from this should form at least part of the conclusion in the abstract.

INTRODUCTION 1. The scope of the review is curable STIs therefore I suggest the detail on non-curable STIs could be removed (lines 75- 77; 78; 81;96) 2. Could "et al" be changed to "et al." (line 97) 3. I think it would be helpful to describe the findings of the review conducted by Terris-Prestholt (2006) as this forms essential background to the current work. 4. Could the authors add a reason why they consider it important to update the previous review (re-phrase sentence starting line 100)?

MATERIALS AND METHODS 1. I was not able to find the protocol described in line 108. Could this be added as an appendix? 2. I was also unable to find Appendix A referred to in line 120. 3. Could you add the initials of the authors who undertook the methods described (line 131;132)? 4. It would be helpful to include more detail on the "cost terms" and extracted cost data (line 135 onwards). The study combines both unit costs and cost-effectiveness measures and it would be helpful to the reader if more detail could be given on these terms. I note that detailed definitions are provided in the work of Terris-Prestholt (2006). 5. The author describes their approach for inflating prices to 2015 value (line138-143). Is there a reference for this method? Why was this method chosen in preference to the method used when the country of origin was not known? (Do the authors have an assumption about a change in the relationship between local currency and US dollars over the time period?) 6. The authors extracted all cost data presented in papers, which includes some not "empirically derived by the authors". I assume this includes parameters included in models? I have a concern about the validity of this, given the strict date criteria of the systematic review. I am interested to know what the approach was to cost data that came from a source publication from prior to 2006 and what date was attributed to its value for the inflation process. 7. I don't agree with the statement in line 162 that the different models "explore the significance of cost effectiveness outcomes versus unit costs". It would be helpful to include more detail on the choice of variables included in the statistical model. 8. The authors have included empiric cost data and cost data from model outputs in the same regression model. Would it be possible to stratify the analysis by the method used to obtain the cost data? 9. Similarly, costs from the provider perspective were combined with costs from the societal perspective. Would it be possible to stratify the analysis by the perspective of the extracted cost value? 10. I cannot comment on the use of robust standard errors or natural log of cost values.

Table 1: This table includes all the included studies therefore I suggest the title is changed from "selected literature" to "included literature". And clarify that year refers to year of publication (not study).

RESULTS 1. Figure 1 is a clear description of the systematic review findings. 2. I think there is a typo on line 177 (45 articles) 1. One paper included costs from a Sub-Saharan country and a Latin American country and looks to have been included in the SSA country category. Could this decision be justified? One paper did not specify a country, could the authors explain how this met the inclusion criteria for the study? 2. The authors describe that "modelling papers" tended to have "larger sample sizes". I assume these were hypothetical population sizes? I would prefer to see the description of the included studies separated into empiric and modelling studies. 3. It would be helpful to provide more detail on the "range of other populations" line 195 as they constitute 10% of the sample. 4. I suggest the authors remove the word "just" from line 196. 5. I think the manuscript would be improved by more description of the included costs (line 231 onwards). The statement in line 234 to interpret the median cost with caution would benefit from more explanation/rationale. For example what are the general patterns in the cost data by for example location or STI or perspective? If word count is a concern perhaps the summary description of the cost data in the text (line 198) could be presented as a table? 6. I would like to read more detail or analysis of the cost drivers as I believe this would be informative for policy makers.

Reviewer #3: The paper aims to update an earlier published systematic review about costs of diagnosis and treatment of STI in low and middle income countries. The authors use the methods from the earlier publication by Terris-Prestholt et al (2006). They https://mail.google.com/mail/u/0/?ui=2&ik=0f32d53aae&jsver=FOnR4BGjAPw.en.&view=pt&msg=15fba55100fe25e6&q=csfirnhaber%40gmail.com&qs… 2/3 2/12/2018 Gmail - Your Submission search the literature published between 2006 and 2014 and include 45 papers into the review. From these papers they extracted cost values and perform statistical analysis on those values. They present results in a large table with information on the individual studies and in some figures and tables with descriptive analysis of the cost values.

While the information in table 1 may be useful, all other results in the figures and tables are in my view not interesting, because they compare numbers that are not comparable. I am not a health economist, so maybe cannot judge the value of the presented analysis fully. Nevertheless, to me it makes no sense to pool all costing values into one data set and provide descriptive analysis. For example, on page 10 the authors give percentages of costing values with certain characteristics (e.g. percentage from sub-Saharan African countries, percentage from health providers perspective). To me these percentages seem completely arbitrary, depending on how many values a paper reports, and on how many papers report certain types of costs. Does it make sense to simply count cost values?

It is also not explained what the authors mean by "unit costs". They seemed to have combined costs together with cost- effectiveness measures like DALYs into the same data set, numbers which are not at all comparable.

On the top of page 11 they talk about a bottom-up approach of calculating costs versus top-down and give percentages again for what part of the cost values was derived in one or the other way. But what insight do we get from these percentages? The authors do not give an explanation or interpretation.

In the section "Cost outcomes" on page 11 they compute a median of all cost values (including all types of cost values into one calculation). In my view these numbers do not make any sense.

On page 12, the authors conclude that "full costs were significantly higher than incremental costs", again a conclusion that does not seem meaningful to me.

Finally, I wonder why the authors limit their time period to 2014. It is now 2017, so why not update their search and analysis to a more recent date?

Reviewer #4: The study updates a previous systematic review and is designed to summarize recent literature on the costs of diagnosing and treating curable diseases. For the period 2006 to 2014 the authors identify 44 articles reporting a total of 202 cost values. These costs values include cost per STI case treated, per patient, per specific infection, per DALY averted, cost for VALacyclovir per patient per month, per patient on first line therapy for First antifungal treatment, per woman tested, per visit, per case of syphilis averted, per HIV infection averted, per tablet, per service provided, per woman screened, per case of PID averted, patient correctly treated, and numerous others. And these were derived using a wide variety of methods.

There are several major problems with this work. 1. Many of the reported results seem to be meaningless. For example the abstract and paper report the median of the 202 cost values. The cost distribution simply comes from the choice of which, and how many, cost values were reported by different researchers addressing a wide range of completely different research questions. What does this median of this distribution mean and how could it possibly be useful? This is just as true for the medians of cost values stratified by cost-effectiveness measures and unit costs. Others include: a. Cost figures resulting from cost effectiveness analyses were nearly 14 times higher than unit costs. b. Costs presented as full costs were 41.35 times larger than incremental costs. These just show that if you are estimating completely different parameters you get completely different answers. In any case, each of these subgroups are themselves so heterogeneous that I can't see what one would do with this information 2. The regression methods are unclear. I presume this is a linear (meta-)regression model on the log-transformed costs? What does 'We have used a linear representation of all but one of the independent variables' mean? It is unclear whether the results are from a multivariable meta-regression model, or several univariable meta-regression models. You have not reported standard errors or confidence intervals for the regression coefficients or made any mention of the level of uncertainty around them. You don't account for the fact that many of the cost values are correlated. ie several cost values are reported for the same patients.

There may be some useful information in the dataset that the authors have compiled but they need to give more thought to what it is. And only report statistics that are calculated using a sound methodology, and that have an understandable and useful interpretation.

https://mail.google.com/mail/u/0/?ui=2&ik=0f32d53aae&jsver=FOnR4BGjAPw.en.&view=pt&msg=15fba55100fe25e6&q=csfirnhaber%40gmail.com&qs… 3/3 Conflict of interest Naomi

AUTHOR DECLARATION FORM

At submission, EVERY AUTHOR listed in the manuscript must READ and COMPLETE the following statements on: (A) Authorship Responsibility, (B) Authorship Criteria, (C) Authorship Contribution, (D) Funding Disclosures, (E) Contributor Disclosures/Acknowledgments, and (F) Conflicts of Interest Disclosures.

It is important that you return this form as early as possible in the publication process. EVERY AUTHOR MUST COMPLETE AN INDIVIDUAL COPY OF THE FORM, AND EVERY SECTION OF THE FORM MUST BE COMPLETED. We will NOT consider your manuscript for publication until every author has completed the form and returned it to us.

Your name (please print): Naomi Lince-Deroche E-mail: [email protected]

Journal name: PharmacoEconomics Corresponding author: Naomi Lince-Deroche

Manuscript title: The costs of diagnosing and treating sexually transmitted infections in low- and middle-income countries

from 2006 to 2014: An updated systematic review

A. AUTHORSHIP RESPONSIBILITY I certify that ALL of the following statements are correct (PLEASE CHECK THE BOX). • The manuscript represents valid work; neither this manuscript nor one with substantially similar content under my authorship has been published or is being considered for publication elsewhere (except as described in the manuscript submission); and copies of any closely related manuscripts are enclosed in the manuscript submission; AND • For manuscripts with more than one author, I agree to allow the corresponding author to serve as the primary correspondent with the editorial office and to review and sign off on the final proofs prior to publication; or, if I am the only author, I will be the corresponding author and agree to serve in the roles described above. • For manuscripts that are a report of a study, I confirm that this work is an accurate representation of the trial results.

B. AUTHORSHIP CRITERIA To fulfil all of the criteria for authorship, every author of the manuscript must have made substantial contributions to ALL of the following aspects of the work: • Conception and planning of the work that led to the manuscript or acquisition, analysis and interpretation of the data, or both; AND • Drafting and/or critical revision of the manuscript for important intellectual content; AND • Approval of the final submitted version of the manuscript. I certify that I fulfill ALL of the above criteria for authorship (PLEASE CHECK THE BOX).

C. AUTHORSHIP CONTRIBUTION I certify that I have participated sufficiently in the work to take public responsibility for (PLEASE CHECK 1 OF THE 2 BOXES BELOW): Part of the content of the manuscript; OR The entire content of the manuscript.

D. FUNDING DISCLOSURES PLEASE CHECK 1 OF THE 2 BOXES BELOW: I certify that no funding has been received for the conduct of this study and/or preparation of this manuscript; OR I certify that all financial and material support for the conduct of this study and/or preparation of this manuscript is clearly described in the Compliance with Ethical Standards section of the manuscript.

Some funding organizations require that authors of manuscripts reporting research deposit those manuscripts with an approved public repository. Please check here if you have received such funding.

E. CONTRIBUTOR DISCLOSURES All persons who have made substantial contributions to the work reported in the manuscript (e.g. data collection, data analysis, or writing or editing assistance) but who do not fulfill the authorship criteria MUST be named with their specific contributions in the Acknowledgments section of the manuscript. Groups of persons who have contributed may be listed under a heading such as ‘Clinical investigators’ and their function described. Because readers may infer their endorsement of the manuscript, all persons named in the Acknowledgments section MUST give the authors their written permission to be named in the manuscript.

I certify that all persons who have made substantial contributions to this manuscript but who do not fulfill the authorship criteria are listed with their specific contributions in the Acknowledgments section in the manuscript, and that all persons named in the Acknowledgments section have given me written permission to be named in the manuscript.

F. CONFLICT OF INTEREST DISCLOSURES A conflict of interest exists when professional judgment concerning a primary interest (such as patients’ welfare or the validity of research) may be influenced by a secondary interest (such as financial gain or personal rivalry). A conflict of interest may arise for authors when they have a financial interest that may influence – probably without their knowing – their interpretation of their results or those of others. We believe that to make the best decision on how to deal with a manuscript we should know about any such conflict of interest that the authors may have. We are not aiming to eradicate conflicts of interests – they are almost inevitable. We will not reject manuscripts simply because the authors have a conflict of interest, but we will publish a declaration in the manuscript as to whether or not the authors have conflicts of interests.

All authors MUST complete the following checklist:

Category If you have had any of the listed relationships with an entity that has a financial interest in the of potential subject matter discussed in this manuscript, please check the appropriate “Yes” box below and conflict of provide details. If you do not have a listed relationship, please check the appropriate “No” box. interest When completing this section, please take into account the last 36 months through to the foreseeable future. No (√) Yes (√) Details Employment √

Grant received/grants √ pending

Consulting fees or √ honorarium

Support for travel to √ meetings for the study, manuscript preparation or other purposes

Fees for participation in √ review activities such as data monitoring boards, etc

Payment for writing or √ reviewing the manuscript

Provision of writing √ assistance, medicines, equipment or administrative support

Payment for lectures √ including service on speakers bureaus

Stock/stock options √

Expert testimony √

Patents (planned, √ pending or issued)

Royalties √

Other (err on the side √ of full disclosure) Every author MUST complete option 1 or option 2 as appropriate below. If you answered "Yes" to any of the questions relating to financial conflicts of interests in the table above (or if you wish to disclose a non-financial conflict of interest), you MUST write a suitable statement in the box below and include this statement in the Compliance with Ethical Standards section of the manuscript.

I have no conflicts of interest to declare; OR

The following statement regarding conflicts of interest and financial support for conduct of this study and/or preparation of this manuscript is to be published in the Compliance with Ethical Standards section of the manuscript:

Declaration: I certify that I have fully read and fully understood this form, and that the information that I have presented here is accurate and complete to the best of my knowledge.

Your name (please print): Naomi Lince-Deroche

Signature (please HAND-WRITE):

Date:12 February 2018

Conflict of Interest Form - Rahma

Conflict of Interest Form - Sharon

'lducsnueul eql ut paueu oq ol uorssruilad ueuu/v\ eu, ue^r6 e^ell uorl9es sluau6pal/v\ou)pv aql ur paLueu suosJad lle ]eq] pue 'ldu3snueur aLll ur uorlces sluau6pall ou1ov aLll ul suorlnqlrluoo ogpeds leql L,llri^ pelsrl are eualuo dtqsloqpe aql llglnl lou op oq^ 1nq lducsnueur srql ol suorlnquluo3 leriuelsqns apeu e^eLl oqrir suos:ad lle leqt i{1qpac I ffi

'1dr:csnueru eql ur paueu aq o1 uorssrured uaprrm.reql sroqlne eql enr6 lsnru uorlcas sluauOpa;rnoupv aq1 ur paLUeu suosrad g;e 'lducsnueu aql lo lueurasropua Jreql Jalur {eu srepea.r asnecag poquosap uor}ounl Jraq} pue ,srole6rlsanur lecrurl3, se qcns 6utpeaq e rapun palstl aq {er"u palnqr4uoc a^eLl oqnn suoslad 1o sdnorg }duosnueLU oq} Jo uor}ces slueLu0pa;rnou)pv eql ur suorlnqrJ}uoo cgrceds raql qlrm psueu aq IS1W erJaluc drqsroqlne aql llgln1 lou op oq^^ 1nq (ecuelsrsse Outltpa ro 6utlt:m ro 'sts,{;eue elep 'uoqcelloc elep '6'a) lducsnueu eq1 ur pepodel IJo/v\ eql ol suorlnqrJluoc lerluplsqns apeu.r aneq oqm suosLed ;;y s3unso']0sto uornEtulNo3':l '6urpun; q3ns pe^recel aneq no,{ }r ereq l1ceqc oseeld n fu oysodar alqnd panordde ue Lllr/v\ sldr.rcsnueu asoql Isodap q3reasa.r 6urpodar syducsnueu ]o sJor.llne 1eq} arrnbar suorlezrue6ro 6urpunl eLuog 'lducsnueu aql Jo uorlces spJepupls leclLlll qlrrvr acuer;duoO aql ur paqrrcsap ,{lrea;c sr lducsnueu srql ]o uorleiederd ro7pue {pn1s srql ,o }cnpuoc aq} ro} poddns leua}eu pue lercueu1 lle }eq} ,ltUac t X y6 llducsnueur srql lo uorlerederd ro/pue {pngs srql lo }cnpuoo aql Jo, pa^ra3ar uaaq seq 6urpunl ou }ellt &tUac t X :MO-l:18 SSXO8 Z f Ht JO I vC3H3:lsv:l']d sSunso-t3sto 0NtoNn|o 'ldricsnueuu eql ]o lueluo3 ejrlue eqt XI UO :ldlFsnueu eql lo lueluoo eql Jo UBd [] :(Mo-]:tE s3xo8 yC3HC Alrpqrsuodsar ur paledrcqed Z IHI JO l 3SV3'Id) ro; l;qnd alel ol lro^ aql {;luarcggns a^eq I }eq},llpac I NOtrn8tuINoS dtHsuoHl-nv'3

.rol ''l-'lv l{1lUec (XOe f Hf y33HC 3svl'ld) drqs:oq1ne eueluc e^oqe eql Jo lluln] I teql t '1dr:csnuelu tr oql lo uorsJe^ pe$ruqns leuU aLll 1o lenorddy OA/y :lualuoo len]oellelur ]ueuodur .rol ldlrcsnueu elll Jo uorsr^al lecryJo Jo/pue 6urgerg ONy :Llloq to 'elep elil lo uorlela.rdralur pue srsl;eue 'uorirsrnbce ro lducsnueu eLll ol pal leql lro/v\..sr.l].1"^9:iir"ld pue uorgdecuo3 . .Iru/vr dL{l }u SiSedSE outanollOi eqi ;o 'l-1v ol suollnquluoc lBrluelsqns apeu a^eq lsnur lducsnueur eql jo roqlne fuana 'drqsroqlne lol BueJu3 aq] to lle lUlnl oI vluSl_tuc drHsuoHrnv'€t 'sllnsal lerJl eql ,o uorleluasardar alerncce ue sr IJo/r slql leLll [lJ4uoc I '{pn1s e 1o podar e eJe }eqi slducsnueu io3 'enoqe peqrJosap soloJ aql ur eruas ol eel6e pue Joqlne Ourpuodsalroo aL{l aq 1lrvr I ':oq1ne {;uo aql ue I Jr 'Jo :uolleo;qnd og.ror.rd s;oord leu4 aql uo go u6rs pue /v\alnar ol pue ecuJo leuolpo eql qll/vl luepuodsaroc fueulr.rd aql se srues ol roqlne Ourpuodse.r:oc or,.ll r olle ol aar6e | 'Joqlne auo ueql eJour r.{}r/v\ slducsnueu ro3 . ONV :uorssruqns lducsnuer-u eql ur pasolcue a;e slducsnupl! pa]ela.r ,{1eso;c lue;o seldoc pue l(uorssrr-uqns lducsnueu aql ur paquosap se ldacxe) ereqa,resle uorlecr;qnd Joj peJeprsuoc 6utaq sr Jo paLlsrtqnd uaaq seq dtqs.roqpe Au "rapun lualuoc leirurs ,{11er1ue1sqns Liir/n auo .rou lducsnueu srr.l} rol-iirau :Iro/v\ prleA siuasa:dar lducsnueu aq1 (XOA yOf sluauralels 6urano11o1 aql ,{1pac HO lSVS-ld) lcaroc are Jo 1eq1 1 :Hf llv x.V AII']ISISNOdS3U dIHSUOHINV

crleLuals{s palepdn uV :VIOZ ol g00Z LUoJ} ^^ar^al selrlunos euo3ur-alpprru pue -/v\ol ul suollcalur pallil.usuel] {1;enxas 6urlea:1 pue 6ursou6erp }o s}so3 aqI :al}r} lducsnueyl

sqcolaO-aourl r.rroeN :loLllnE 6urpuodsalo3 ssrurouoclocBr.uJeqd :ot!eu leujnol

uuoc 1reLu6@e16ea1 zue :peiu-f e;6ea1 earpuv :(1ur:d esee;d) aueu .rno1

sn ol lr pouJnleJ pue uiJol aq1 paleldiuo3 seLl rorllne fueae ;ryun uorlecrlqnd ro; lducsnueu lno,{ reprsuoc ION lll/$ eM 'O3I31dtllO3 :lS ISflW WUOJ :lHI lO NOI133S AU3A3 ONV 't IUOJ ]Hr lO ,\dOS ']VnOnlONl NV 3lf-ldt/!lo3 'ssacord ISnUU UOHI1V AUfnJ uo4ecr;qnd aq1 ur alqrssod se ,{pea se ulro} srr{i urn}oJ noI 1eq1 }ueuodtur sr }l

'saJnsol3sr6 lsaJalul ,o slorlluoC (l) pue 'sluauOpalmoulcv/seJnsolcsrq lolnquguo3 (3) 'sarnsolcsrC 6urpun3 (6) 'uor1nqr4uo3 drqsroqlny (3) 'eua1u3 drqsroqlny (g) 'Alrlrqrsuodsa5 drqsLoqlnv (V) :uo sluauielels 6urmo;1o; eql 31:t-1dt41g3 pue OVIU lsnLu lducsnueu aLl] ur palsrl UOHI1V AUSnf 'uorssruqns ]V

SIPV V IAIUOJ NOIIVUV]C3O UOHINV Conflict of Interest Form - Andrea - Form Interest of Conflict (aLnsolcsrp 11n1 1o eprs aql uo la) laqlg

serle{og

(panssr Lo Durpuad 'peuueld) slualed

,{uourrlsel Uadxf

suorldo )colsplcols

snearnq steleads uo egrruas 6urpnlcur seJn)3al lo; 1uatu,{e6

poddns eAtlellstutupe rc luaudrnba 'saur3rpaul'ecuelstsse 6urlurvr;o uotstnold

lducsnueu.r aq1 6urrrnarnat ro 6urlunn:o1 luaur{e6

cla'spJeoq Ouuolruour Blep se qonS S3lll^llCe,v\elAeJ ur uoriedrcr;led io1 sae3

sasodlnd laLllo Jo uorleredatd lducsnueu 'Ipn1s aq1 lo1 s6urleer.u o] la^eJl to; l.roddng

urntJeJouoLl Jo seel 6ur11nsuo3

6urpued sluer6Tpanrecor lueJO

luau{ogdtu3 sllBloo (P) se,l (p) ot-t 'aJnlnj elqeeeseJo] ,uotlces eq1 o1 q6no.tq] sqluour g0 lsel aql lunocce olut aIeI esea;d slql Ourlelduroc uarl n lssJalu! 'xoq elerrdordde eL{} paq3 'drqsuorlela.r ,,oN,, eseald pe}sll e a^eq }ou op nod }l slre}3p aprnoid Jo lcllluos pue olaq xcq ,,se^,, aleudotdde aqi )3eq3 aseald '}dr:csnueu stql ut possncstp lalleu ^ ]cahns le4ue1od 1o aL{} ur }saJa}ut letcueuu e seq leq},{1r}ua ue q1r,rn sdrqsuor}elar pa}stl aql 1o r{ue peq aneq no{1; tuoEe1e3

:lstllceqc Dutanollol eq1 elelduoc ISnW sJoqlne llv 'slselalut slsluuoo e^eq sroqlne lo aq} lou Jo Jeqlaq/\^ o1 se ldtJcsnuetx aql ul uorlBJeleap e qsrlqnd lltM a/\^ lnq 'lseralut Jo lotuuoc e a^eq sJoLllne alll esne3aq I;dtuts s$ucsnueu 'elqe]t^aut 1cafer ]ou lil/s eM ]soujle ale ,{eq1 - slseja}ur Jo s}cruuoc aleclpera o1 oulLute lou eJe aM e^eq {etu sroqlne aql }eql }seje}ut Jo }otlluoo qcns ,{ue lnoqe n oul plnoqs em ldr.rcsnueut qll/v\ e leop ol Moq uo uolsl3ep lsaq eLll a)eu ol leLll a^olleq aM SJeLllo ]o esoql .lo Sllnsal Jlaql Jo uorleletd.ralur lraql * 0utmoul lloql lnoLllt^ Ilqeqord - aouanBut IeLu ieql ]saralur lercueuU e a^pq {eq1 uaqan sroq}ne Jo} espe Ier,u 1sela1ur Jo lcl[uo3 y (fu|ent: leuostad to ure6 letcueuq se qcns) lsaJalut fuepuoces e ,{q pecuangul aq ,{eu (qc.rease:1o ,{1pr;en aql Jo qcns) ale]la^ ,s1uat1ed se lsstalut fueuud e 6uturecuoc yuatu6pnt;euorssalotd uaq/v\ slstxa Isalalut ]o lctlluoc V sSunso-t3sto rs3uSlNt lo r3t-'HNoc'J

}\ 8LoZ tuenrqal tL:alec

:(fIlUM-ONVH aseatd) ernleu6rs

a16ea1 eeJpuv :Quud eseald) aureu tnol

a6palMoul [Lu 1o lsaq aq] o] elalduoo pue a]Blnooe st elaLl paiueserd a^Bq I leql uorleurolur aql leql pue 'trrol srql poolsrapun ,{;1n; pue pear {;1n; eneq ; 1eq1 ,{;rpac | :uotleJeloag

:]du3snueLu aql jo uotl3es sprepuels lecllllf qlrnn ecuer;duroo aq] ur peqsrlqnd aq ol st ldr.rcsnuetu srqy 1o uorlelederd toTpue ,{pn1s srq} }o }enpuoc .ro1 iroddns leroueur} pue }sero}ur }o s}eluuoc 6urp.re6et }uarrla}E}s 6urano1lol eql_ E UO :erel3ap ol lsalalur lo s]oilluoc ou a^eq I X 'lducsnueu eLll lo uotlcas sptepueis leclqlf qltan ecuetldtuoC aql ul lueuralEls srql epnlcur pue Molaq xoq eLll ur luauralels alqeuns e alt.lM 19661 no{ 'Qsatalut jo lotguoo letcueug-uou e asolcsrp ol rlsrn^ no,{;r Lo) a^oqe elqel aq} ut s}saJe}ut }o s}otlJuoc let3ueuq o} Ourlelet peieMsue suotlsanb aql 1o ,{ue o} ,,sa^,, nor( 11 anolaq aleudoidde se 7 uorldo to 1. uorldo alalduoc 1g1-1t^l roqlne fuan3

\ Conflict of Interest Form - Calvin

Conflict of Interest Form - Cindy